Nelson Garcia-Vazquez, Shaoren Yuan, Moustafa Gabr bioRxiv - Pharmacology and Toxicology DOI: 10.64898/2026.02.21.707154 Abstract Protein-protein interactions (PPIs) mediated by extracellular ligands remain challenging targets for small molecule intervention due to their large and dynamic interfaces. The interaction between SLIT2 and its receptor ROBO1 plays a critical role in cell migration and tumor progression, yet remains largely unexplored. Here, we report the discovery and optimization of small molecule inhibitors of the SLIT2/ROBO1 interaction enabled by DNA-encoded library (DEL) screening. Affinity selection against SLIT2 identified four structurally diverse hit compounds, which were subsequently validated using orthogonal biophysical assays. Among these, one hit exhibited measurable SLIT2 binding and functional inhibition of the SLIT2/ROBO1 interaction in a time-resolved FRET assay. Guided by physicochemical considerations, a solubility-optimized analog was designed, resulting in a ~50-fold improvement in binding affinity and an ~9-fold enhancement in functional potency. Molecular dynamics simulations and induced-fit docking revealed a stable binding mode within the SLIT2 LRR2 domain and suggested that a benzothiophene substituent was dispensable for target engagement. Fragment-based experimental validation confirmed this prediction, leading to the identification of a minimal azaindole-based pharmacophore that retained nanomolar binding affinity. Collectively, this study demonstrates how DEL-enabled hit discovery combined with rational optimization and fragment deconstruction can yield potent small molecule modulators of a challenging extracellular PPI, providing a foundation for further development of SLIT2/ROBO1 pathway inhibitors.

Evan S. O’Brien, Junzheng Wang, Parthasaradhireddy Tanguturi, Mengchu Li, Elizabeth White, Yuki Shiimura, Barnali Paul, Kevin Appourchaux, Kaavya Krishna Kumar, Weijiao Huang, Susruta Majumdar, John R. Traynor, John M. Streicher, Chunlai Chen, Brian K. Kobilka bioRxiv - Biophysics DOI: 10.64898/2026.02.20.707058 Abstract Opioid agonists such as morphine and fentanyl exert analgesic effects by binding and activating the µ-opioid receptor (µOR), yet agonism of the µOR causes a slate of serious side effects. µOR-mediated addiction and respiratory depression are the major causes of the current opioid overdose crisis, largely driven by the explosion in illicit use of fentanyl, a potent opioid receptor full agonist. Given these serious side effects (and high resulting societal cost), molecules that act as analgesics with distinct mechanisms of action are of great interest. Positive allosteric modulators (PAMs) of the µOR have the potential to avoid many off-target side effects of conventional opioid orthosteric agonists by enhancing the signaling properties of natural opioid peptide systems. We used a DNA-encoded chemical library screening approach to selectively discover active-state-specific µOR PAMs. Two out of 3 selected prospective PAMs displayed the anticipated enhancement in agonist activity. The most effective of these compounds enhanced the activity of all orthosteric opioid agonists tested, including the native opioid peptide met-enkephalin. Little is known about the underlying dynamic basis of allosteric modulation of Family A GPCRs like the µOR. To that end, we used single-molecule fluorescence resonance energy transfer experiments to detail the impact that our novel µOR PAM has on the dynamic activation behavior of a key region on the intracellular face of the receptor. Our results here provide both a new chemical scaffold that acts as a µOR PAM and detailed pharmacological and dynamic insights into its mechanism of action.

Jia-ying Xue,Jia-hui Shi,Yuan Yao,Wei-en Xie,Yong Zou,Ming Yan,Xue-jing Zhang Organic Letters DOI: 10.1021/acs.orglett.6c00234 Abstract We report a skeletal editing strategy based on DNA-encoded nitroarenes for the direct conversion of benzene cores into valuable 3H-azepine scaffolds. This transformation is efficiently promoted by visible light in the presence of P(Oi-Pr)3, which serves as a reductant to generate reactive nitrene intermediates from the nitro group. Demonstrating broad substrate scope with applicability to pharmaceutical molecules, this protocol offers an efficient and versatile route to DNA-encoded 3H-azepine derivatives. It thus establishes a robust platform for skeletal diversification in DNA-encoded library synthesis.  

Congbao Kang ,  Hung T. Nguyen ,  David E. Heppner ,  Bin Yu ,  Weijun Xu Journal of Medicinal Chemistry DOI: 10.1021/acs.jmedchem.6c00070   Over the past decade, RNA has emerged as an attractive and evolving landscape for small-molecule drug discovery. RNA functions as an intermediate macromolecule during gene expression and plays a diverse role in regulating cellular processes. (1) The promise in targeting RNA as therapeutic interventions arises from recent breakthroughs in structural biology, chemical biology, and computational modeling that collectively facilitate understanding of RNA biology. Multiple classes of RNA have been identified contributing to their functional diversity, including messenger RNA (mRNA), transfer RNA (tRNA), ribosomal RNA (rRNA), long noncoding RNA (lncRNA), microRNA (miRNA), and other noncoding RNAs (ncRNAs). While only a tiny fraction of 1.5% of human genome is translated into proteins, approximately 75% is transcribed into RNA, (2−4) making RNA a vastly larger reservoir of potential drug targets than the traditional protein targets. Moreover, many disease-associated pathways including viral replication, cancer progression, neurodegeneration, and immune regulation are regulated by RNA or through its interactions with different proteins, making RNA an increasingly relevant and strategic focus for next-generation drug discovery. (5) Several classes of small molecules define the landscape of RNA-targeting drug discovery (Figure 1), and many other modulators of RNA have been discovered for further development. (6) Small molecules targeting RNA can act through diverse mechanisms, including direct RNA binding to modulate RNA function, disruption of RNA interactions with RNA binding proteins, and stimulation RNA binding to RNA degrading enzyme for degradation. (7) The success in developing RNA drugs and RNA ligands prove the feasibility of RNA-focused drug discovery. (2) Patients with spinal muscular atrophy (SMA) lack enough survival motor neuron (SMN) protein to maintain adequate muscle function. The lack of SMN protein is believed to drive the pathophysiology of SMA. In August 2020, the FDA approved risdiplam (Evrysdi, RG7916, RO7034067) developed by Roche for the treatment of SMA in patients of all ages. (8) As a small-molecule modulator of SMN2 splicing, risdiplam represents the first FDA-approved RNA-targeting small-molecule drug. Mechanically, risdiplam acts as a “glue”-like compound that enhances RNA binding protein (RBP) to RNA to facilitate splicing and increases the production of full-length functional SMN protein. In addition to risdiplam, there are other drugs or clinical candidates that directly bind to RNA, including small molecules targeting RNA G-quadruplexes such as Quarfloxin and CX-5461, ribosome-targeting antibiotics like macrolides, and viral RNA polymerase nucleosides like sofosbuvir. (9) Together, these precedences highlight the growing potential of RNA as a therapeutic target and lay the foundation for the rational discovery of small-molecule RNA modulators. Figure 1. Small-molecule drugs that target RNAs. While strong incentives have been directed to drug RNA, several fundamental challenges make RNA-targeted drug discovery inherently more intractable than conventional protein-directed approaches. (10) Although proteins can undergo complicated conformational changes and post-translational modifications, RNA is even more dynamic and flexible, adopting distinct conformations that depend on its sequence, cellular environment, and interactions with other biomolecules. (11) Said another way, the highly dynamic nature of RNA implies that robust pockets may be impossible to bind small molecules like unstructured regions of proteins. Unlike proteins formed by 20 amino acids, RNA is composed of only four nucleotides, presenting challenges for specific and tight interactions with small molecules. Additional hurdles include the limited identification of clearly druggable RNA motifs, the intrinsic structural dynamics of RNA, and the difficulty in achieving selectivity across closely related RNA sequences. Additionally, traditional protein-directed drug discovery often relies on both polar and nonpolar interactions within well-defined hydrophobic and hydrophilic pockets. By contrast, RNA features overwhelmingly negatively charged backbones and typically lacks hydrophobic pockets. As a result, the number and structural diversity of druggable RNA binding sites are expected to be limited, which further complicates efforts to achieve selective modulation. As such, the structural and functional nature of RNA intrinsically favors chemical space dominated by highly polar molecules manifesting to at least two major issues: limited selectivity due to the abundance of similarly polar features across cellular RNAs and significant challenges in optimizing ADME properties, particularly lipophilicity. As with drug discovery of protein targets, hit identification via high-throughput screening (HTS), fragment-based screening, and virtual screening are commonly adopted. A major challenge for HTS against RNA targets is that most existing HTS libraries were designed for proteins and are poorly suited for recognizing the unique physicochemical and structural characteristics of RNA. Therefore, the development of RNA-focused libraries has become a key priority, which requires pharmacophore-based strategies through incorporating scaffolds known to interact with RNA motifs. This can be further complemented by the rise of DNA-encoded libraries (DELs) that have emerged as powerful source for hit identification for RNA targets. (4) For RNA to become a more tractable and widely druggable class of biomolecules, advances are needed in generating well-behaved biochemical tools and obtaining detailed structural information on unique or transient RNA conformations. Such innovations will be essential for enabling robust discovery platforms analogous to those that have long supported successful drug development for proteins and enzymes. Despite the above-mentioned challenges, RNA-targeted therapeutics continue to gain momentum across oncology, virology, and neurology. Recent advances in structural biology and computation, artificial intelligence, and multiscale modeling have begun to overcome these barriers and pave a rational foundation for RNA-targeted drug discovery. Structural studies of RNA using chemical biology and biophysical methods have made progress in recent years, providing detailed insights into secondary and tertiary motifs such as hairpins, internal loops, bulges, and pseudoknots, which are considered important structural architectures for developing small molecules. NMR spectroscopy remains a powerful technique for probing RNA structure and dynamics in solution. (12) Its ability to resolve conformational equilibria, detect transient states, and characterize ligand binding has made it indispensable for understanding RNA functional motions and identifying druggable conformations. X-ray crystallography, usually hindered by the intrinsic flexibility of RNA, has become increasingly feasible due to improved construct design, presence of stabilizing ligands, and the use of RBPs to facilitate crystallization. These innovations have enabled high-resolution structures of diverse RNA motifs and RNA–ligand complexes to be solved. (13) Cryo-EM has transformed the field through efficient determination of large RNA molecules and RNA–protein complexes that were previously inaccessible. (14) Complementing experimental approaches, computational methods have evolved rapidly and played a critical role in RNA structural biology and drug discovery. (15) Several algorithms and servers have been developed to predict RNA secondary and tertiary structures, model conformational ensembles, binding pocket identification and screen compound libraries against RNA targets. (16) Machine learning based methods that are supported by continuously expanding high-quality structural data, have improved accuracy in predicting RNA structures, RNA–RBP interactions and RNA–ligand interactions. Collectively, these methods provide a robust foundation for rational discovery of RNA modulators. Advances in physics-based molecular dynamics (MD) simulations are now mapping RNA conformational landscapes to reveal hidden and metastable pockets. Enhanced sampling MD, (17) coarse-grained (CG) modeling, (18) and Markov state models (MSMs) (19) allow the identification of transient states that are invisible to standard spectroscopy and experimental determination. (20) Coupled with ensemble-based docking against multiple conformers, hit rates may improve compared to static docking against RNA targets. Furthermore, the integration of hybrid QM/MM and machine-learning (ML) corrected scoring functions is providing a more rigorous treatment of stacking interactions, hydration shells, and ion-mediated electrostatics, which are critical for accurate RNA-ligand affinity prediction. These methods are increasingly used to derive druggability maps for riboswitches, repeat RNAs, viral elements, and structured motifs within long non-coding RNAs. (21) On the other hand, AI-driven prediction and generative design for RNA-targeted chemistry is on the rise. However, a central barrier is the limited availability of high-resolution RNA–ligand structures, which constrains traditional structure-based design. Emerging AI approaches are addressing this gap through multimodal learning frameworks that integrate RNA sequence, chemical features, SHAPE/DMS reactivity, and evolutionary covariation. Deep learning models improve secondary and tertiary structure inference, enabling more accurate identification of ligandable motifs and conformational states relevant for binding. (22) Recently, prediction of small-molecule–RNA interactions was achieved without the need for RNA tertiary structures as input. (23) However, their broader utility in RNA-focused medicinal chemistry remains to be fully established and will likely require larger data sets and more rigorous experimental validation. Moreover, many disease-relevant RNAs function through multivalent interactions with RBPs. Small molecules that restore normal RNA–RBP equilibrium have demonstrated proof-of-principle activity in correcting splicing, transcriptional regulation, and translation. (24) It remains to be seen whether disrupting RNA–protein binding interaction would bring clinical benefits. As data sets from CLIP–seq, RNP–MaP, and ligand–RNA cross-linking expand, predictive modeling of RBP selectivity will continue to improve. Last but not least, a rapidly growing therapeutic direction targets ribonucleoprotein condensates, that feature dynamic assemblies whose physicochemical properties (viscoelasticity, aging kinetics, etc.) encode key regulatory functions. (25) Aberrant condensate behavior is implicated in many diseases including neurodegenerative diseases and cancer. (25) Small molecules that alter condensate properties, such as softening, hardening, dissolving, or preventing gelation, represent a new modality of RNA-focused therapeutics. (26) Computational modeling is expected to play a central role capturing the structures of mesoscale organization and emergent behaviors of ribonucleoprotein condensates. These approaches collectively support rational design of condensate-modulating drugs, a modality fundamentally different from classical lock-and-key pharmacology. Overall, deep understanding of RNA targets, innovative screening and validation as a collaborative effort from chemical biology and medicinal chemistry is key to the success of RNA targeted drug discovery campaigns. With rapidly advancing techniques in structural biology and artificial intelligence are being developed, we may witness whether small molecule drug discovery targeting RNAs will meet its inflection point or remain a niche curiosity in the years to come. This article references 26 other publications. (acccessed 2024/10/22) PubMed (acccessed 2026/01/04) (acccessed 2026/01/04) This article has not yet been cited by other publications.

Xianfeng Li ,  Zehao Yin ,  Qiuyi Chen ,  Xinlong Hu ,  Gong Zhang ,  Xiaohong Fan ,  Yizhou Li Organic Letters  DOI: 10.1021/acs.orglett.6c00490 Abstract The β-ketoamide motif represents both a privileged scaffold and a versatile synthetic intermediate in medicinal chemistry. Herein, we developed a DNA-compatible method for the efficient conversion of various DNA-conjugated amines into β-ketoamides. The resulting β-ketoamides facilitate rapid diversification into a panel of structurally diverse molecular scaffolds. Importantly, the synthetic route and subsequent derivatization steps were validated to be fully compatible with DNA encoding, offering a reliable and versatile platform for DNA-encoded library synthesis.  

Yulong An ,  Ruolan Zhou ,  Xiang Li ACS Medicinal Chemistry Letters  DOI: 10.1021/acsmedchemlett.5c00738 Abstract DNA-encoded library (DEL) technology has emerged as a transformative platform for discovering chemical inducers of proximity (CIPs), addressing challenges in both degrader and non-degrader CIP development. This Microperspective analyzes the results of recent DEL technology screens (2021–2025) to enable medicinal chemistry programs, focusing on CIP development including CIP-focused DELs, DEL-derived ligands for proteins of interest (POIs) and E3 ligase in rational CIP design, and directly functional CIP identification. Finally, we address current limitations of DEL technology in CIP research and outline future directions. This Microperspective underscores DEL’s pivotal role in advancing CIP discovery, providing actionable insights for addressing “undruggable” targets and accelerating translational research in chemical biology and medicinal chemistry. Summary This MicroPerspective reviews recent advances (2021–2025) in DNA-encoded library (DEL) technology for discovering chemical inducers of proximity (CIPs), spanning both degraders (e.g., PROTACs, molecular glue degraders) and non-degrader modalities (e.g., protein stabilization, subcellular relocalization, transcriptional activation). It synthesizes three key strategies: (1) CIP-focused DELs (CIP-DELs), enabling simultaneous dual-target (POI + E3 ligase) selection to directly identify cooperatively binding bifunctional compounds; (2) Conversion of DEL-derived POI/E3 ligands—leveraging well-defined DNA attachment sites as “exit vectors”—into functional CIPs; and (3) Discovery of non-degradative CIPs, including FKBP12-recruiting molecular glues and function-driven DEL screening (e.g., direct ubiquitination readout). DEL overcomes longstanding limitations of traditional HTS—including library size, cost, and scarcity of E3 ligands—thereby accelerating CIP development against “undruggable” targets. Highlights Dual-Target CIP-DEL Screening: CRBN- or VHL-targeted DELs enable concurrent selection against POIs and E3 ligases, directly identifying ternary complex stabilizers with high cooperativity (e.g., BRD4/BRD2-selective PROTACs, BRD9 molecular glue). Ligand-to-CIP Conversion Paradigm: DEL-derived ligands for ERα, MAGE-A3, PIN1, DNPH1, and TRIM21 were optimized and converted into functional PROTACs or TrimTACs; the DNA attachment site serves as a built-in, precise “exit vector” for linker conjugation. Expansion to Non-Degradative Functions: An FKBP12-biased CIP-DEL identified a molecular glue that stabilizes the Crohn’s disease-associated ATG16L1 T300A variant; function-driven DEL screening (in presence of E1/E2/ATP) directly enriches ubiquitination-competent PROTACs, eliminating affinity-only false positives. New Frontier: RNA Targets: DEL screening against RNase L led to the design of RiboTACs targeting pre-miR-21—extending CIP therapeutics to RNA biology. Conclusion DEL technology has evolved from a single-target ligand discovery platform into a central engine driving the discovery of the full spectrum of CIPs—from degraders to non-degraders, and from proteins to RNA. Its core advantages lie in vast chemical space coverage, barcode-enabled precise hit identification, and intrinsic structural information (e.g., defined exit vectors). Future directions include integrating AI for POI–E3 interface–guided library design, developing robust in-cell DEL screening, expanding the repertoire of E3 ligase ligands, and strengthening functional phenotypic and preclinical translational studies—to fully unlock the therapeutic potential of CIPs against “undruggable” targets.

Qingao Xue ,  Ze Liang ,  Yi Zhang ,  Fei Wang ,  Fulian Wang ,  Lili Liu ,  Guang Yang ,  Lei Yan Bioorganic Chemistry DOI:10.1016/j.bioorg.2026.109627 Abstract The membrane fusion process mediated by the SARS-CoV-2 spike protein is a key therapeutic target. Its heptad repeat 1 (HR1) domain forms a conserved trimeric groove critical for forming the fusion-competent six-helix bundle with HR2. We used DNA-encoded library screening to identify small molecules binding HR1. Hits including Rabeprazole-related compound E (Rab RCE), Omeprazole, Alvimopan, and Olmesartan were characterized. Biophysical assays confirmed binding, while computational simulations revealed distinct interaction modes, with Alvimopan showing high predicted affinity. Cell-cell fusion assays demonstrated potent inhibitory activity for Olmesartan and Rab RCE. Notably, Rabeprazole and Rab RCE showed partial antiviral activity against SARS-CoV-2 variants and HCoV-OC43, rescuing virus-induced apoptosis. Mechanistically, Rabeprazole competitively occupies the HR2-binding groove on HR1, blocking fusion. Our findings identify HR1-targeting molecules like Rabeprazole as promising leads for broad-spectrum coronaviral fusion inhibitors. Highlights A DNA-encoded library (DEL) screening strategy was established to rapidly identify small-molecule binders targeting the conserved heptad repeat 1 (HR1) domain of the SARS-CoV-2 spike protein, enabling efficient mining of repurposed drug candidates from a ∼ 4 billion-compound chemical space. Four clinically approved drugs (alvimopan, olmesartan, rabeprazole sulfide, and omeprazole) were validated as HR1-targeting agents, sharing biaryl/heteroaryl cores and hydrogen-bond acceptor groups that mediate specific interactions with HR1 (binding affinities ranging from micromolar to millimolar). Two distinct inhibitory mechanisms were delineated: classical competitive occupancy of the HR1 hydrophobic groove (olmesartan, rabeprazole sulfide) and a novel ‘molecular wedge’ mode disrupting the trimeric HR1 interface (alvimopan), providing complementary strategies for targeting viral fusion. Olmesartan and rabeprazole sulfide exhibited potent inhibition of SARS-CoV-2-mediated cell-cell fusion, with efficacy comparable to the positive control Salsingle bondC, validating their potential as lead compounds for anti-COVID-19 therapeutics. This study establishes a robust pipeline integrating DEL screening, biophysical validation, molecular docking, and functional assays, offering valuable chemical scaffolds and mechanistic insights for developing broad-spectrum coronaviral fusion inhibitors.  

Kejia Yan ,  Guilherme M. Lima ,  Tara Bahadur ,  Vincent Albert ,  Zoe O’Gara ,  Gary Bao ,  Christin Kossmann ,  William Kirby ,  Fernando B. Mejia ,  Matthew L. Michnik ,  Kristen Maiorana ,  Ratmir Derda bioRxiv - Biochemistry DOI: 10.64898/2026.02.14.705946 Abstract Genetically encoded (GE) libraries enable identification of high-affinity ligands for diverse molecular targets through iterative in vitro selection and DNA sequencing or next-generation sequencing (NGS). Despite their impact in therapeutic development, a systematic framework for evaluating reproducibility in GE-molecular discoveries remains limited. To aid such analysis, we introduce the concept of baseline response, which reproducibly partitions active and inactive members of in vitro selection. The baseline response is provided by spiking a random DNA-barcoded population. We calibrated the baseline concept using Bioconductor EdgeR differential enrichment (DE) analysis of NGS of phage-displayed selection on oligosaccharide chitin and hepatitis virus NS3a* protease as model targets. We further show that mixing discovery campaigns also offers an effective baseline: chitin-enriched peptides serve as a baseline for DE-analysis of NS3a* selection and NS3a*-enriched peptides serve as a baseline for chitin binders. We applied baseline-stratified DE-analysis to 66 parallel selections performed in 3–5 replicates across 22 extracellular targets, including HER1-3, EpCAM, CAIX, PD-L1, and eight integrin receptors. Automated DE-analysis across hundreds of NGS files produced hits validated in a secondary screen and yielded synthetic macrocyclic ligands with mid-nanomolar affinity confirmed in 2–3 biophysical assays. For PD-L1, we further demonstrated how baseline-calibrated NGS data provide decision-enabling information for optimization of peptide macrocycles to yield potent single-digit nanomolar ligands for the cell-surface receptor. We anticipate that baseline-based analyses of NGS data from in vitro selection procedures will offer a scalable framework for reproducible hit discovery and standardized analysis across diverse in vitro selection campaigns. Summary This work introduces a universal baseline framework for in vitro selection of genetically encoded (GE) libraries—e.g., phage-displayed peptide libraries—to improve reproducibility, statistical rigor, and cross-target comparability. The core innovation is spiking a DNA-barcoded random peptide library (serving as an in situ or “cross-target” empirical baseline) into every selection round. This baseline mimics naïve library binding behavior and enables robust normalization and differential enrichment (DE) analysis using Bioconductor EdgeR on NGS data. Validation spanned 22–24 extracellular protein targets (including HER1–3, PD-L1, integrins, NS3a*, chitin) across 66 parallel selections. Baseline-stratified DE identified high-confidence hits, including synthetic macrocyclic ligands with mid- to single-digit nM affinity confirmed by biophysical assays. The method also supports functional benchmarking—e.g., revealing reduced infectivity in MBX-modified phage libraries—and replaces synthetic or computational baselines with empirically derived, target-agnostic mixtures. Highlights Spiked DNA-barcoded random peptides serve as composition-agnostic, in situ baselines for normalization. Cross-target library mixing (e.g., chitin + NS3a*-selected peptides) yields effective empirical controls. EdgeR-based DE with TMM normalization and BH-FDR correction (α = 0.05) enables quantitative FC estimation binned by input abundance. Baseline depletion <1% after NS3a* selection confirms high selectivity. Conclusion The universal baseline standardizes hit discovery, improves enrichment fidelity assessment, and enables ML-ready, statistically benchmarked data generation without structural priors.

Yoojin Kim ,  Hayeon Kim ,  Jinhui Hong ,  Minseo Kang ,  Jaeyoung Bae ,  Sangyoon Ko ,  Minjae Kim ,  Byumseok Koh ,  Hakjoong Kim ,  Sang-Hee Shim ,  Kyubong Jo bioRxiv - Bioengineering  DOI: 10.64898/2026.02.15.706034 Abstract DNA-encoded library (DEL) technology enables high-throughput small-molecule discovery but is typically performed using purified proteins under in vitro conditions that do not reflect native intracellular environments. Here, we present a microfluidic agarose microdroplet platform for cellular-context DEL screening. The porous hydrogel droplets provide mechanically stable yet permeable microenvironments that protect weak protein-ligand interactions while enabling efficient buffer exchange and ligand diffusion. Importantly, mild cell permeabilization within droplets selectively retained chromatin-associated proteins, allowing screening directly in a cellular context. Using BRD4 as a model target, we validated intracellular ligand engagement by fluorescence imaging and super-resolution microscopy. Small-scale DEL screening selectively enriched JQ1 in both bead-based and cell-based formats, and large-scale DEL screening across millions of encoded compounds successfully identified hit molecules by sequencing. This agarose microdroplet based strategy expands DEL technology toward biologically relevant and chromatin-associated targets under near-native conditions. Summary This work introduces a microfluidic agarose µ-droplet platform for DNA-encoded chemical library (DECL/DEL) screening against intracellular targets, with validation on BRD4—a nuclear epigenetic reader protein. The system encapsulates single cells or target-coated magnetic beads in monodisperse ~100 µm agarose droplets via flow-focusing microfluidics; the low-gelling-temperature (LGT) agarose forms a porous hydrogel upon cooling, permitting rapid diffusion of DNA-encoded small molecules while preserving intracellular architecture and protein–bead complexes. Permeabilization enables controlled probe access, and super-resolution Exchange-PAINT imaging confirms nanoscale colocalization of JQ1-BP with GFP-BRD4 in nuclear nanoclusters. Highlights Agarose µ-droplets enable gravity-assisted washing, shear protection, and uniform molecular diffusion. Two-color Exchange-PAINT with orthogonal R2/R6 docking strands validates specific intracellular target engagement. DEL screening yields target-specific enrichment: JQ1 barcode is selectively enriched in BRD4-overexpressing HeLa droplets. Scalable to large DELs (96×96×96 combinatorial space across three scaffolds) with nanopore sequencing–based enrichment quantification. Conclusion The platform bridges functional intracellular DEL screening with high spatial fidelity and quantitative readouts—enabling both PCR- and sequencing-based hit identification while preserving native biomolecular context.

Maxwell Kleinsasser ,  Brayden J. Halverson ,  Edward Kraft ,  Sean Francis-Lyon ,  Sarah E. Hugo ,  Mackenzie R. Roman ,  Ben Miller ,  Andrew D. Blevins ,  Ian K. Quigley arXiv - QuanBio - Biomolecules  Abstract The quality and consistency of training data remain critical bottlenecks for protein-ligand binding prediction. Public affinity datasets, aggregated from thousands of labs and assay formats, introduce biases that limit model generalization and complicate evaluation. DNA-encoded chemical libraries (DELs) offer a potential solution: unified experimental protocols generating massive binding datasets across diverse chemical and protein target space. We present Hermes, a lightweight transformer trained exclusively on DEL data from screens against hundreds of protein targets, representing one of the largest and most protein-diverse DEL training sets applied to protein-ligand interaction (PLI) modeling to date. Despite never seeing traditional affinity measurements during training, Hermes generalizes to held-out targets, novel chemical scaffolds, and external benchmarks derived from public binding data and high-throughput screens. Our results demonstrate that DEL data alone captures transferable protein-ligand interaction representations, while Hermes' minimal architecture enables inference speeds suitable for large-scale virtual screening. Summary The paper introduces Hermes, a lightweight transformer-based model trained exclusively on DNA-encoded library (DEL) screening data across 239 protein targets. Despite never using traditional affinity measurements (e.g., IC50, Kd), Hermes generalizes to unseen protein targets, novel chemical scaffolds, and external benchmarks derived from public binding data. The model demonstrates that DEL data alone captures transferable protein-ligand interaction representations, with inference speeds 500–700× faster than state-of-the-art structure-based models like Boltz-2, making it highly suitable for large-scale virtual screening. Highlights Strong generalization: Achieves mean AUROC of 0.68 on the DEL Protein Split (unseen proteins) and 0.60 on Public Binders/Decoys (external benchmarks), with significantly better performance for kinase targets due to kinase-enriched training data. Speed advantage: Processes 28.2 samples/second/GPU on H200 hardware, far outpacing Boltz-2 (0.04 samples/second on H100), critical for cost-effective virtual screening. Limitations: Performance drops on the DEL Chemical Library Split (AUROC ~0.56), suggesting challenges in generalizing to entirely new chemical libraries. Data binarization (binary binding labels) and noise in DEL screening results constrain model expressivity. Practical impact: Highlights DEL datasets as a scalable, unified alternative to fragmented public affinity data (e.g., ChEMBL), with potential to accelerate drug discovery pipelines. Conclusion Hermes demonstrates that DEL-derived data alone can train generalizable protein-ligand binding prediction models without reliance on traditional affinity measurements. Its success underscores the value of large-scale, consistent DEL screening data for capturing transferable biological interactions. As DEL datasets continue to grow beyond public affinity resources, DEL-trained models like Hermes are poised to drive the next generation of computational drug discovery, particularly for targets underrepresented in existing public data. Future improvements could incorporate structural augmentation and continuous binding strength modeling to address current limitations.

Amanda W. Dombrowski，Florent Samain ACS Medicinal Chemistry Letters DOI: 10.1021/acsmedchemlett.6c00047 Abstract Over the past 30 years, the field of DNA-encoded libraries (DELs) has become a mature and robust technology platform for the identification of ligands against relevant biological protein targets. Most of the major innovative pharmaceutical companies have integrated DEL platforms into their drug discovery workflows. Indeed, DELs have significantly impacted drug discovery efforts in the last 10 years with the identification of ligands that have progressed into clinical trials for various disease indications. One could assume that there are likely even more DEL-derived ligands that have reached the clinic, but candidate stories do not necessarily mention the hit generation methods utilized. (1−3) The fundamental concept of DELs emerged in a theoretical paper from Lerner and Brenner in 1992. (4) DELs are defined as collections of small molecules that are covalently attached to unique DNA tags, serving as amplifiable identification barcodes. Encoding procedures allow the generation and screening of combinatorial libraries of high diversity and unprecedented size. Preferential binding molecules identified by high throughput DNA sequencing of libraries after affinity capture procedures are typically resynthesized and tested to characterize their binding properties. (5) Success of DEL-based drug discovery screening campaigns is affected by various factors, including quality and diversity of the libraries, screening protocols, protein selection conditions, downstream validation and data analysis methods. Advances in the field have expanded DEL applications beyond traditional synthesis and screening procedures. Recently, the incorporation of artificial intelligence (AI) and machine learning (ML) techniques in DEL workflows has been accelerating, driving significant advancements and extending the potential of technology. (6,7) This ACS Medicinal Chemistry Letters Collection features recent success stories that highlight the value and impact of DEL technology in drug discovery. The following report, A Novel Small Molecule Allosteric Inhibitor of IL-17A from a DNA-Encoded Library, demonstrates the ability of DEL to identify small molecules that directly modulate the IL-17 pathway by inhibiting IL-17A with their cognate receptors, proving that small molecules can mimic the action of macromolecular biologics to disrupt high affinity protein–protein interactions. DEL macrocycle-like libraries have also proven effective for the recognition of larger protein surfaces (DNA-Encoded Macrocyclic Peptide Libraries Enable the Discovery of a Neutral MDM2–p53 Inhibitor). Beyond traditional inhibitor identification, DEL technology has become a pivotal approach for the development of targeted protein degradation (TPD) therapeutics. The report, Discovery of Small-Molecule Ligands for the E3 Ligase STUB1/CHIP from a DNA-Encoded Library Screen, from AstraZeneca scientists, shows that DEL can be effectively used to identify small molecule ligands for STIP1 homology and U-box containing protein 1 (STUB1), an E3 ligase that contains protein–protein interaction (PPI) sites, where previously only peptide binding ligands have been discovered. DEL hits can also serve as tools to provide structural basis for further hit-to-lead progression, thus accelerating medicinal chemistry lead optimization activities (Structural and Molecular Insight into the PWWP1 Domain of NSD2 from the Discovery of Novel Binders Via DNA-Encoded Library Screening; Optimization of a Novel DEL Hit That Binds in the Cbl-b SH2 Domain and Blocks Substrate Binding; Chemical Space Profiling of SARS-CoV-2 PLpro Using DNA-Encoded Focused Libraries). By leveraging the broad chemical diversity offered by DEL techniques, researchers have been able to identify robust covalent inhibitors that specifically interact with cysteine residues. This targeted approach helps improve both the selectivity and efficacy of potential therapeutic agents (Identification and Evaluation of Reversible Covalent Binders to Cys55 of Bfl-1 from a DNA-Encoded Chemical Library Screen). Recent literature indicates that DEL designs have become more strategic, influenced by progress in synthetic methodologies, expanded chemical diversity, and enhanced access to structural biology data. The perspective, Design of DNA Encoded Libraries for Medicinal Chemistry, provides a comprehensive analysis of DEL derived hits that enabled medicinal chemistry programs from publications between 2020 and 2025. The development of computational tools has offered new opportunities to rationalize DEL designs and DEL data set analysis. The integration of DEL technology with computational approaches, such as machine learning, continues to unleash the potential of the technology (Highly Selective Novel Heme Oxygenase-1 Hits Found by DNA-Encoded Library Machine Learning beyond the DEL Chemical Space; Evaluating the Diversity and Target Addressability of DELs using Scaffold Analysis and Machine Learning). This Collection features 10 publications, which highlight the rapid growth in many areas of the DEL field. We thank the authors for their work, which we hope will inform and inspire our readers to expand upon that work. This article references 7 other publications. This article has not yet been cited by other publications.   Summary This editorial introduces a special collection of 10 publications highlighting the rapid evolution and maturation of DNA-encoded library (DEL) technology over the past 30 years. DELs, which consist of small molecules covalently attached to unique DNA barcodes for identification, have transitioned from a theoretical concept (first proposed by Lerner and Brenner in 1992) to a robust, industry-standard platform for drug discovery. Major pharmaceutical companies have now integrated DEL platforms into their workflows, with several DEL-derived ligands progressing into clinical trials over the last decade. The editorial emphasizes that recent advances—including AI/ML integration, novel library designs, and expanded applications beyond traditional inhibitor discovery—are significantly extending the potential of this technology. Highlights 1. Diverse Therapeutic Applications IL-17A Inhibition: DEL technology successfully identified small molecule allosteric inhibitors that disrupt high-affinity protein-protein interactions (PPIs), demonstrating that small molecules can mimic macromolecular biologics. Macrocyclic Libraries: DEL macrocycle-like libraries enabled the discovery of neutral MDM2-p53 inhibitors, effective for recognizing larger protein surfaces. Targeted Protein Degradation (TPD): DEL screening identified small molecule ligands for STUB1/CHIP (an E3 ligase containing PPI sites), where previously only peptide binders were known. 2. Structural and Mechanistic Insights DEL hits provide structural foundations for hit-to-lead optimization, accelerating medicinal chemistry efforts. Examples include: Novel binders to the PWWP1 domain of NSD2 Inhibitors targeting the Cbl-b SH2 domain Chemical space profiling of SARS-CoV-2 PLpro 3. Covalent Inhibitor Discovery Strategic DEL designs enable the identification of reversible covalent binders targeting specific cysteine residues (e.g., Cys55 of Bfl-1), improving selectivity and efficacy. 4. Integration of Artificial Intelligence and Machine Learning AI/ML techniques are increasingly incorporated into DEL workflows for: Rationalizing library designs Analyzing DEL datasets Discovering hits beyond traditional DEL chemical space (e.g., highly selective heme oxygenase-1 inhibitors) Scaffold analysis and target addressability prediction 5. Strategic Library Design Evolution Recent DEL designs have become more sophisticated, influenced by: Progress in synthetic methodologies Expanded chemical diversity through validated on-DNA chemical reactions Enhanced access to structural biology data Conclusion The editorial concludes that DEL technology has become a mature, robust, and indispensable platform in modern drug discovery. The featured collection demonstrates the technology's versatility across multiple applications—from traditional inhibitor identification to targeted protein degradation and covalent drug discovery. The integration of computational approaches, particularly machine learning, continues to unlock new potential and extend DEL capabilities beyond its original scope. The editors (Amanda W. Dombrowski and Florent Samain) express gratitude to the contributing authors and anticipate that these works will inform and inspire readers to further expand the boundaries of DEL technology in pharmaceutical research.

Vasanthanathan Poongavanam ,  S. Pauliina Turunen ,  Kristian Sandberg ,  Ulrika Yngve ,  Johan Wannberg Drug Discovery Today  DOI: 10.1016/j.drudis.2026.104629 Abstract DNA-encoded libraries (DELs) combined with machine learning (ML) offer a powerful paradigm for hit identification. However, sequencing-derived enrichment data are inherently noisy and biased, often resulting in models that overfit to specific chemical libraries. In this review, we critically evaluate the capabilities and limitations of DEL-ML, illustrating key challenges using Aurora Kinase A (AURKA) DEL affinity selection data. We demonstrate that standard ML models often struggle to generalize to unseen chemical space because of the specific structural constraints of combinatorial libraries. Furthermore, we discuss the necessity of rigorous denoising strategies and evaluate approaches, such as domain adaptation, to mitigate these limitations, offering a roadmap for building robust models capable of exploring diverse chemical space. Summary This review critically examines the integration of machine learning (ML) with DNA-encoded library (DEL) technology for drug discovery. While DEL-ML offers a powerful paradigm for hit identification by generating massive binding datasets (10⁶–10¹² data points), the authors identify a critical "generalizability gap" that limits the practical utility of current models. Using Aurora Kinase A (AURKA) as a case study with OpenDEL 4.0 screening data (~1.5 million data points), the authors demonstrate that standard ML models achieve high accuracy on internal validation but frequently fail to generalize to structurally novel scaffolds due to domain shift—the substantial difference between DEL chemical space and known pharmacological compounds. The review provides methodological best practices for data preprocessing, denoising, and validation, while evaluating advanced strategies such as domain adaptation to improve model robustness. The authors argue that future DEL-ML development must move beyond simple accuracy maximization toward explicit handling of distribution shifts to transform DEL-ML from a retrospective analysis tool into a reliable engine for novel chemical discovery. Highlights 1. The Generalizability Challenge in DEL-ML Models trained on DEL data often memorize library-specific building blocks rather than learning transferable structure-activity relationships The BELKA competition revealed that models perform well on test sets within the same chemical space but fail on structurally novel scaffolds Domain shift between DEL training data and external compound collections represents a fundamental barrier to practical application 2. Data Quality and Preprocessing Considerations DEL sequencing data contains unique noise profiles including matrix binding, DNA-tag interference, unequal synthesis yields, and "jackpot" effects Multiple denoising strategies are evaluated: fold-enrichment, Z-scores for ultra-large libraries, disynthon aggregation, and uncertainty-aware probabilistic loss functions Critical importance of subtracting background noise from control experiments (matrix/bead-only) to prevent false positives 3. Class Imbalance and Data Splitting Strategies DEL selections produce highly imbalanced datasets (10¹–10⁴ binders vs. up to ~10¹² nonbinders) Random splitting leads to overoptimistic metrics due to high structural similarity within DEL congeneric series Scaffold-based or library-based splitting provides more rigorous assessment of generalizability to novel chemotypes Undersampling nonbinders (e.g., 1:1 ratio) can boost external sensitivity from ~1% to 20–30%, though this may reflect bias exploitation rather than true generalization 4. Molecular Representation and Model Architectures Traditional fingerprints and physicochemical descriptors often fail to capture subtle variations in DEL compounds Graph neural networks (GNNs) and variational autoencoders (VAEs) show promise but require careful handling of linker/DNA-tag artifacts Compositional (disynthon) approaches reduce sparsity but risk losing "whole-molecule" structural fidelity Conformal prediction frameworks provide calibrated confidence intervals essential for prioritizing predictions in noisy DEL environments 5. Domain Adaptation as a Solution Strategy Covariate shift correction reduces divergence between source (DEL) and target (known binder) domains Using high-confidence predictions from diverse compound collections (e.g., Enamine REAL Diversity Set) as an intermediate domain improves generalization Domain adaptation reduced PCA centroid distance from 0.77 to 0.32 between DEL training data and known AURKA space Retraining with both predicted binders and nonbinders improved Matthews Correlation Coefficient (MCC) from 0.2 to 0.4 on external datasets while maintaining 20–39% sensitivity 6. AURKA Case Study Findings OpenDEL 4.0-derived binders tended to be larger, more lipophilic, and less polar compared to known AURKA inhibitors Despite overall domain shift, highly enriched DEL hits from sublibrary 27 shared conserved hinge-binding motifs with established inhibitors (e.g., VX-680) Mechanistic alignment between DEL hits and known binders confirms that domain shift, rather than fundamental binding mode differences, drives prediction failures Conclusion The integration of DELs with ML presents transformative opportunities for early drug discovery, but realizing this potential requires overcoming the critical generalizability gap. The primary challenge is not data volume but data nature: intrinsic structural biases and systematic false negatives (often linker-induced) cause models to memorize library-specific artifacts rather than learn transferable pharmacophore principles. High internal validation metrics frequently mask failures to extrapolate to novel, pharmacologically relevant scaffolds. The authors advocate for a paradigm shift in DEL-ML development emphasizing: Rigorous validation standards: Moving beyond random splits to scaffold-based and out-of-distribution evaluation Domain alignment strategies: Explicit handling of distribution shifts through domain adaptation and transfer learning Data diversity expansion: Open-source DEL datasets spanning broader drug-like chemical space to reduce single-library bias Integration of physics-based priors: Incorporating docking constraints to reduce overfitting to synthetic artifacts Uncertainty quantification: Systematic use of conformal prediction and applicability domain assessment By pivoting from simple accuracy maximization to robust domain alignment, DEL-ML can evolve from a retrospective analysis tool into a reliable engine for identifying novel chemical starting points. The establishment of standardized benchmarks and community resources will be essential to accelerate the development of generalizable predictive models capable of exploring the vast chemical space beyond individual DEL compositions.

Saan Voss , Amin Sagar , Arnaud Tiberghien , Richard J. L. Hughes , Liuhong Chen , Inmaculada Rioja , Mark Frigerio , Michael J. Skynner , David R. Spring Journal of Peptide Science DOI: 10.1002/psc.70071 Abstract Bicyclic peptides are emerging as next generation therapeutics by combining the affinity and specificity of antibodies with the synthetic convenience of small molecules. Phage-encoded libraries of bicyclic peptides enable the discovery of high-affinity molecules against virtually any protein target. The generation of bicyclic peptides that advanced into clinical development involves the reaction of three cysteines in a peptide to a C3-symmetric alkylating agent. In phage display, this chemical modification transforms a pool of conformationally flexible peptides into a library of structurally unique protein mimetics that are able to bind traditionally challenging protein surfaces like those with limited structural definition. In recent years, a new class of bicyclic peptides has emerged using a single atom-bismuth-in place of C3-symmetric organic scaffolds, thus expanding into an unexplored chemical space at the intersection of inorganic chemistry and biology. This mini-review aims to reflect on the discovery, evolution and potential future applications of bismuth bicycle molecules.

Christian M. Gampe , Bigna Wörsdörfer , Ge Zou , Antonio Ricci ACS Medicinal Chemistry Letters DOI: 10.1021/acsmedchemlett.5c00676 Abstract Despite advancements in hit-finding technologies, many drug targets are considered difficult-to-drug (D2D) or difficult-to-ligand (D2L). Here, we present an analysis of 21 hit-finding campaigns across three research organizations within the Roche group, focusing on D2D and D2L targets. DNA-encoded library technology (DELT) was the most successful method in providing validated hits and lead series. High-throughput, covalent, and peptide screens also yielded progressable chemical matter in a substantial number of cases. In contrast, fragment and virtual screens, while effective in generating validated hits, demonstrated lower success rates. Stratifying targets into D2D and D2L categories provided a useful framework for estimating the likelihood of project success and informing additional screening strategies, with D2D targets showing higher rates of chemical enablement. Our findings indicate DELT as a valuable experimental tool for assessing ligandability and highlight the importance of informed integrated hit discovery by tailoring hit-finding strategies to target characteristics.

Yagong Wang , Huanqing Zhang , Fanming Zeng , Xue Zhao , Junyun Chen , Lijun Xue , Kexin Yang , Yun Jin Hu Chemistry - An Asian Journal DOI: 10.1002/asia.70611 Abstract Dithiocarbamates (DTCs) are privileged scaffolds in medicinal chemistry, yet inaccessible via DNA-encoded libraries (DELs) due to a lack of robust on-DNA synthesis. We developed a general procedure for on-DNA DTC formation using a carbon disulfide (CS2) bridging strategy. This method efficiently links diverse aliphatic secondary amines and alkyl halides under mild conditions with high conversions and excellent DNA compatibility. The utility of this method was demonstrated by constructing a prototype DEL, thereby bridging a critical gap in chemical space and facilitating the rapid discovery of DTC-based therapeutics.

Sangeeta Pandey , Florent Samain , Omprakash Nacham , Jon D. Williams , Nathaniel L. Elsen Expert Opinion on Drug Discovery DOI: 10.1080/17460441.2026.2622373   Abstract INTRODUCTION Affinity selection mass spectrometry (AS-MS) is a powerful label-free technique for characterizing macromolecule-ligand interactions that has been used as a hit finding tool with significant success. Recent advances in MS and separation technology have positioned AS-MS to impact more areas of drug discovery. AREAS COVERED This manuscript provides a brief historical review of AS-MS and the recently developed technologies that have enabled AS-MS. The report also provides examples and references for how AS-MS has been used for high-throughput screening (HTS) to DNA-encoded library (DEL) screening hit confirmation, Direct-to-Biology, and natural product screens. The references for this work were collected from a broad range of sources, including Google Scholar, Scopus, review articles identified via Google Scholar, and the internal AI resource at AbbVie Inc. EXPERT OPINION AS-MS is a unique biophysical binding assay that does not rely on labels and can specifically detect binders from large pools of potential ligands based on molecular weight. There is still significant room for growth in areas of impact that will be driven by decreases in separation time and a move toward equilibrium conditions during separation. Increased use for driving rapid structure-activity relationships (SAR) has potential to decrease project cycle times in lead identification and optimization.

Francesco V Reddavide , Trine L Toft-Bertelsen , Ieva Drulyte , Aspen Rene Gutgsell , Dzung Nguyen , Sara Bonetti , Katerina Vafia , Anne-Sophie Tournillon , Stephan Heiden , Grosser Grosser , Katarina Iric , Veronica Diez , Nanna MacAulay , Stefan Geschwindner , Thompson Thompson , Jens Frauenfeld , Robin Loving bioRxiv - Biochemistry DOI: 10.64898/2026.01.27.701919 Abstract Developing novel drugs against membrane proteins is a major challenge in drug discovery due to the difficulty of stabilizing these targets for high-throughput screenings. Pannexin 1 (PANX1) is a membrane channel protein involved in various physiological and pathological processes, making it a promising target for drug discovery. However, efforts to develop PANX1-targeting therapeutics have been hindered by the inherent challenges of stabilizing the protein channel and conducting effective pharmacological screening. Here, we report a proof-of-concept workflow that integrates the Salipro lipid nanoparticle platform with DNA-Encoded Library screenings in a detergent-free format. In this case study, the Salipro DirectMX method was used to generate functional PANX1 nanoparticles for drug discovery and characterisation. Using a high-stringency selection strategy and computational approaches, we identified a specific set of candidate compounds with selective PANX1 enrichment. Surface Plasmon Resonance analysis confirmed the identification of hit compounds. Cryo-Electron Microscopy of the Salipro-PANX1-Compound complex provided structural insights into a potential compound binding site. Electrophysiological recordings in PANX1-expressing Xenopus laevis oocytes demonstrated dose-dependent inhibition of PANX1-mediated ion conductance by the compounds. These findings establish a robust workflow for ligand discovery against challenging membrane protein targets and provide novel chemical starting points for the development of PANX1 modulators.

Baljit Kaur , Longfei Zhang , Hossam Nada , Laura Calvo-Barreiro , Moustafa Gabr RSC Medicinal Chemistry DOI: 10.1039/d5md00943j Abstract Chitinase-3-like protein 1 (CHI3L1, also known as YKL-40) has emerged as a central effector of astrocyte-mediated neuroinflammation and a promising biomarker for Alzheimer's disease (AD). However, small molecule CHI3L1 inhibitors that modulate neuroinflammation are limited. Here, we report the discovery of a CHI3L1-targeted small molecule, DEL-C1, identified through DNA-encoded library (DEL) screening and validated using orthogonal biophysical, computational, and cellular approaches. DEL-C1 demonstrated direct CHI3L1 binding in microscale thermophoresis (MST) and surface plasmon resonance (SPR) assays, with reversible and concentration-dependent association. Molecular docking and 100-ns molecular dynamics simulations revealed a stable binding mode within the CHI3L1 substrate groove, anchored by Tyr206 and flanked by Trp99 and Trp352, supporting a thermodynamically favorable interaction. In vitro ADME profiling indicated a balanced physicochemical profile, permeability, and metabolic stability, consistent with CNS drug-like properties. Functionally, DEL-C1 reversed CHI3L1-induced astrocyte dysfunction by restoring Aβ uptake, lysosomal acidification, and proteolytic activity, while reducing CHI3L1 and IL-6 secretion. DEL-C1 also suppressed CHI3L1-driven NF-κB transcriptional activation, highlighting its anti-inflammatory potential. Collectively, this study establishes DEL-C1 as a promising small molecule modulator of CHI3L1 and a chemical tool to interrogate astrocyte-driven neuroinflammation in AD.

Zsofia Lengyel-Zhand , Zhaowei Jiang , Justin I. Montgomery , Hongyao Zhu , Keith Riccardi , Richard Corpina , Woodrow Burchett , Mario Abdelmessih , Robert Stanton , Timothy K. Craig , Timothy L. Foley  bioRxiv - Biochemistry DOI: 10.64898/2026.01.20.700605 Abstract DNA-encoded library screening represents a significant advancement in the field of drug discovery. Its ability to rapidly and cost-effectively identify potential drug candidates from large compound libraries has the potential to revolutionize the way new medicines are discovered and developed. While the strategies for DEL screening and data analysis have improved over the years, data normalization remains an open challenge. Existing normalization methods can yield poor correlation for compounds with high read count, and they do not account for inherent sources of noise. To overcome these drawbacks, we have developed a robust normalization technique using an antibody fragment and a DNA-conjugated peptide as an internal control. This innovative approach allows for normalization between samples of different conditions and accounts for technical challenges that occur during screening.

Linjie Li , Xudong Wang , Peiqi Ding , Xuanjing Shen , Hangchen Hu , Xiaoxi Wang , Rui Jin , Xinyuan Wu , Yiwei Zhang , Weiwei Lu , Jinfeng Yue , H.Eric Xu , Wanchao Yin , Yechun Xu , Xiaojie Lu JACS Au DOI: 10.1021/jacsau.5c01698 Abstract Allosteric regulation is a central mechanism for modulating biological functions and offers an attractive strategy in drug discovery, particularly for targets considered challenging or “undruggable.” However, the discovery of allosteric inhibitors is hindered by poorly defined binding sites and the lack of effective screening approaches. Here, we present a dual DNA-encoded library (DEL) screening strategy that integrates reversible DEL and covalent DEL (CoDEL) technologies to identify novel allosteric inhibitors of the SARS-CoV-2 RNA-dependent RNA polymerase (RdRp). Using this approach, we discovered the first covalent allosteric inhibitors of RdRp, which engage a previously uncharacterized pocket on the nsp8 subunit and form a covalent bond with Cys114. Subsequent SAR studies and biochemical assays confirmed the allosteric mechanism and elucidated structural determinants of activity. This work highlights the power of integrating reversible DEL screening with CoDEL screening for ligand discovery and establishes a generalizable strategy to identify covalent allosteric modulators for therapeutically important targets for therapy or active probe design.

Jason S. Rush , Joshua D. Wertheimer , Steven D. Goldberg , Donald Raymond , Mateusz Szuchnicki , Andrew J. Baltus , Jeff Branson , Christopher F. Stratton , Aaron N. Patrick , Ruth Steele , Suraj Adhikary , Amanda Del Rosario , Annie Liu , Noah J. Gomersall , Michael Chung , Matthew J. Ranaghan , Xiebin Gu , Marta Brandt , Zhifang Cao , Adrian Bebenek , Blayne A. Oliver , Kasper Hoebe , Lawrence M. Szewczuk , Jennifer D. Venable , Daniel B. Graham , Jennifer Towne , Ramnik J. Xavier Cell DOI: 10.1016/j.cell.2025.12.013 Abstract Human genetic association studies highlight key genes involved in disease pathology, yet targets identified by these analyses often fall outside the traditional definitions of druggability. A rare truncated variant of the scaffold protein CARD9 is linked with protection from Crohn’s disease, prompting us to pursue the development of inhibitors that might similarly modulate innate inflammatory responses. Using a phased approach, we first identified a ligandable site on CARD9 using a structurally diverse DNA-encoded library and defined this site in detail through X-ray crystallography. Building upon this, a subsequent ligand displacement screen identified additional molecules that uniquely engage CARD9 and prevent its assembly into scaffolds needed to nucleate a signalosome for downstream nuclear factor κB (NF-κB) induction. These inhibitors suppressed inflammatory cytokine production in dendritic cells and a humanized CARD9 mouse model. Collectively, this study illustrates a strategy for leveraging protective human genetic variants and chemical biology to tackle challenging targets for dampening inflammation.

Yiran Huang ,  Xiao Tan ,  Xiaoyu Li ,  Feng Xiong ,  Siu Ming Yiu Bioinformatics (Oxford, England) DOI: 10.1093/bioinformatics/btag001 Abstract Motivation DNA encoded library (DEL) technology has been developed as a powerful platform for drug development. Live cell-based selection methodologies were recently developed to expedite drug candidate discovery with higher biological relevance. Nevertheless, hit characterization is challenged by prominent background signals of cell-based selections. Therefore, automated data processing streamline compatible with noisy sequencing output is highly desirable. Results Herein we report an innovative automatic method that enables the most promising hit identification from large quantities of cell-based DEL datasets with improved accuracy and efficiency. This processing workflow is based on a comprehensive unsupervised algorithm incorporating data pre-processing, feature extracting and outlier filtering, descriptor-based classification, similarity score ranking and active compound prediction. We performed methodology development with two DEL selection datasets targeting insulin receptor (INSR) on live cells, from both ˜30 million- and 1.033 billion- membered libraries. The automated scheme has demonstrated high consistency with experimental results as well as self-adaptivity to on-cell DEL datasets with varied library scales. Extended methodology application to cellular thrombopoietin receptor (TPOR) further substantiated the algorithmic generalization capability regarding target proteins. Thus, this approach can serve as a widely applicable workflow automatically differentiating hit compounds and thereby facilitates drug development from candidate discovery.

James H. Walker ,  Kejia Yan ,  Ratmir Derda Biochemistry DOI: 10.1021/acs.biochem.5c00646 Abstract Peptide-derived macrocycles are an emerging class of therapeutics capable of modulating protein–protein interactions that remain inaccessible to small molecules. Genetically encoded library (GEL) platforms such as phage and mRNA display have accelerated macrocyclic ligand discovery by linking peptide sequence to genotype and enabling selections from libraries with up to 1013 members. Efforts to expand the chemical space of GELs have included incorporation of electrophiles, either to generate libraries of true covalent ligands or to enable intramolecular reactions such as peptide cyclization. In the latter case, the electrophile is consumed during library construction, producing transient covalent libraries that enhance stability and diversity but are not designed for direct covalent engagement with targets. By contrast, recent advances have established robust strategies for embedding persistent electrophilic warheads that remain intact during library preparation and selectively react with nucleophilic residues on proteins. These approaches have yielded both reversible and irreversible covalent inhibitors against diverse classes of proteins, while also highlighting challenges in balancing electrophile reactivity with library integrity. Complementary developments in DNA-encoded covalent libraries further underscore the breadth of discovery platforms, though genetically encoded approaches remain uniquely powerful for macrocyclic peptides. Together, these advances define the trajectory of covalent genetically encoded libraries (cGELs) and point toward new opportunities for discovering ligands to historically undruggable targets.

Daniela Schaub , Alice Lessing , Gerlis von Haugwitz , Fabian Meyer , Jörg Scheuermann , Rebecca Buller ACS Central ScienceDOI: 10.1021/acscentsci.5c01516 Abstract DNA-encoded libraries (DELs) have become a powerful platform in drug discovery, practiced both by the pharmaceutical industry and academia. Each small molecule contained in a DEL is covalently linked to a DNA tag which serves as an amplifiable barcode facilitating binder identification. However, the chemical diversity accessible in DELs remains limited by the need to perform reactions under conditions that preserve the integrity of the DNA tag. Additionally, chemical reactions must proceed with high efficiency and selectivity to minimize side products and unreacted starting materials, which cannot be removed and may hamper hit identification. Consequently, expanding the DEL chemical space requires the development of methods that combine high reaction performance with DNA compatibility. In this outlook, we highlight the potential of enzymatic catalysis for on-DNA synthesis, which offers a promising route to expand DEL-accessible chemical space.

John C. Faver , Flora Sundersingh , Lauren A. Viarengo-Baker , Ying-Chu Chen , Katelyn Billings , Patrick F. Riley , Ching-Hsuan Tsai , Christopher S. Kollmann Journal of Medicinal Chemistry DOI: 10.1021/acs.jmedchem.5c02259 Abstract DNA-encoded chemical libraries (DELs) enable the highly efficient screening of billions of small molecules for binding to a target of interest and provide valuable training data for machine learning models for virtual screening. However, DEL screening data are notoriously noisy due in large part to significant variance in the synthetic yield of library members. Here, we show an analysis from a split-sample DEL screening strategy against Bruton’s tyrosine kinase (BTK), which includes a panel of affinity selections against the target at varying concentrations and a probabilistic model to estimate the binding affinity and relative input concentrations of library members. We compared model predictions to SPR measurements of resynthesized DNA-conjugated compounds and found that this methodology yielded an improved ranking of library members by binding affinity compared to enrichment metrics. Additionally, the method successfully recovered a library member with a potent binding affinity that would not have been detected in our standard DEL selection.

Gui-Ping Gao , Quan-Ke Li , Jin-Cheng Ma , Zhi-Jun Zhang , Shao-Yong Zhang , Ying-Qian Liu Bioorganic & Medicinal Chemistry DOI: 10.1016/j.bmc.2025.118541 Abstract Indole, an aromatic heterocyclic compound formed by the fusion of a benzene ring with a pyrrole ring, is widely distributed in the secondary metabolites of plants, animals, and marine organisms. Owing to its unique physicochemical properties and high structural modifiability, indole derivatives can engage in specific interactions with various biological targets, demonstrating a broad spectrum of bioactivities including anticancer, anti-inflammatory, antiviral, and antibacterial effects. Consequently, indole holds an indispensable position in innovative drug discovery and development. This review provides a comprehensive summary of the primary strategies employed in the discovery of indole derivatives. These encompass structure optimization approaches inspired by natural products, such as structure simplification, diversity-oriented synthesis (DOS), biology-oriented synthesis (BIOS), the “pseudo-natural product” (PNP) strategy, and bioinspired synthesis based on biosynthetic building blocks. Additionally, strategies like scaffold hopping, molecular hybridization, drug repurposing, and multicomponent reactions (MCRs) for constructing indole-based molecules are discussed. Particular emphasis is placed on target structure-based discovery strategies for indole derivatives, including ligand-based structure modification, molecular docking-assisted high-throughput virtual screening, and fragment-based drug design (FBDD). Furthermore, the application of emerging techniques such as phenotypic screening, DNA-encoded library (DEL) technology, and free energy perturbation (FEP) calculations in indole-based drug research and development is highlighted. This review aims to systematically organize the multi-dimensional R&D framework for indole derivatives, analyze the specific value of each strategy in addressing drug discovery challenges, and provide a theoretical foundation and methodological support for the rational design and development of novel indole-based drugs. It is anticipated that this work will further enhance the efficiency and innovation level in the development of this class of compounds.

Brent Clayton , Steven M Massey , Shaoyou Chu , Emily R Mason , Stephanie J Bissel , Logan M Bedford , Stacey J Sukoff Rizzo , Andrew D. Mesecar , Bridget L Kaiser , Emma K Lendy , Bruce T. Lamb , Alan D. Palkowitz , Timothy I. Richardson Alzheimer's & Dementia DOI: 10.1002/alz70859_103518 Abstract Background The role of microglia in neuroinflammation is widely recognized as a key contributor to the pathogenesis of Alzheimer’s disease (AD). Genome‐wide association studies have identified PLCγ2 as a key contributor, with specific variants conferring either risk or protection. Notably, the protective PLCγ2•P522R variant is associated with increased mRNA expression, protein levels, and innate activity, whereas the risk variant PLCγ2•M28L shows the opposite trend. Based on these findings, we hypothesize that small molecules capable of enhancing PLCγ2 expression or directly activating the protein could mimic the protective effects of the P522R variant. Such an approach may represent a promising therapeutic strategy to mitigate disease progression and cognitive decline in AD patients. Method We performed high‐throughput screening including DNA Encoded Library (DEL) and Affinity Selection Mass Spectrometry (ASMS) using full‐length protein to identify novel small molecules which bind to PLCγ2. Target engagement was confirmed using Differential Scanning Fluorimetry (DSF) and Cellular Thermal Shift Assay (CETSA). Structure activity relationship (SAR) studies were performed to synthesize analogs and optimize for binding and cellular pharmacology in IP‐One and phagocytosis assays. Top compounds have been studied in vivo to assess pharmacokinetic properties as well as impact on neuroinflammation. Result Novel PLCγ2 activators have been discovered and preliminary optimization has been completed. These compounds have shown positive results for target engagement, biochemical activity, and cellular pharmacology. In silico predictions indicated the molecule structures are suitable CNS drug discovery program starting points. Conclusion Activation of PLCγ2 is a novel therapeutic strategy for treatment of AD. We identified structurally distinct molecular scaffolds capable of enzyme activation and cellular activity. Recommendations for use of probe molecules in target validation studies and the development of lead‐like molecules for clinical studies will be made.

U Hang Chan , Fengling Li , Frances M. Bashore , Scott Houliston , Catherine Vu , Irene Chau , Alison D. Axtman , Levon Halabelian Alzheimer's & Dementia DOI: 10.1002/alz70859_096394 Abstract Background To diversify Alzheimer’s Disease (AD) drug targets, a bioinformatics core is established to provide an unbiased ranking of AD risk‐associated genes by integrating multiple lines of genetic and multi‐omic evidence. From which, several RNA helicases, including RIG‐I‐like receptor 3 (LGP2), melanoma differentiation‐associated protein 5 (MDA5) and Dead Box 1 (DDX1) have been identified as high priority targets differentially expressed in AD brains. All three helicases play a role in the innate immune response pathway against viral RNA. Given the previous link between viral infection and AD pathology, this prompted the development of small molecule chemical probe against these targets to further elucidate their roles in AD. Method Purified proteins were used for ATPase assay development and compound screening. The ATPase assay was performed in the presence of annealed 24mer RNA, double‐stranded RNA (dsRNA) with a 25‐nt 3ʹ overhang, or single‐stranded DNA (ssDNA). We employed DNA‐encoded chemical library (DEL) and computational methods for small molecule hit discovery. Hit confirmation was carried out by ATPase assay, Surface Plasmon Resonance (SPR), Differential Scanning Fluorimetry (DSF) and 19Fluorine‐ Nuclear Magnetic Resonance (19F‐NMR). Hit expansion was carried out for the most promising hits to increase potency and selectivity. Result We describe the development and optimization of a bioluminescence assay to kinetically characterize the activity of three human RNA helicases involved in innate immune response pathway, including MDA5, LGP2, and DDX1. Through DEL‐ML screening, we identified a selective hit for MDA5, and characterized its activity by ATPase assay with IC50 of 8 µM, and orthogonally confirmed by F‐NMR. Ongoing studies aim to elucidate the ligand binding site using X‐ray crystallography. Conclusion We present a robust high‐throughput in vitro assay designed for small molecule screening in a 384‐well format, enabling hit optimization and facilitating the discovery of inhibitors for MDA5, LGP2, and DDX1. Through DEL‐ML screen, we identified a selective MDA5 inhibitor that can be used to further interrogate its role in AD pathogenesis, and serve as a chemical starting point for future drug discovery efforts. This ligand represents first‐in‐class small molecule inhibitor for MDA5, a target that has been underexplored in the context of its role in AD.

Benjamin C. Whitehurst , Niall A. Anderson , Argyrides Argyrou , Peter Astles , Bernard Barlaam , Elaine B. Cadogan , Luca Carlino , Gavin W. Collie , Alex Edwards , Linda Kitching , Yaqin Li , Alexander G. Milbradt , Jenni Nikkilä , Sarah Northall , Sara Pahlén , Saleha Patel , Wendy Savory , Markus Schade , Jonathan A. Spencer , Darren Stead , Christopher J. Stubbs , Aquan Wang , Wenxin Wang ACS Medicinal Chemistry Letters DOI: 10.1021/acsmedchemlett.5c00651 Abstract DNPH1 is a hydrolase enzyme that degrades the noncanonical nucleotide 5-hydroxymethyl-2′-deoxyuridine 5′-monophosphate (hmdUMP), thus acting as a nucleotide pool sanitizer by preventing its aberrant incorporation into DNA. Recent studies have shown that loss of DNPH1 enhances the sensitivity of homologous recombination repair-deficient cancer cells to PARP inhibitors, highlighting its potential as an attractive therapeutic target. Herein we report the design and prosecution of an integrated hit finding strategy combining high-throughput screening, DNA-encoded library screening, and fragment-based lead generation which enabled the discovery of the first non-nucleotide ligands for DNPH1. We compare four hit compounds which differ markedly in their chemical structures, physicochemical properties, and binding modes and summarize parallel hit-to-lead workup efforts. We also provide discussion of the merits of an integrated approach for hit discovery when applied to challenging novel targets such as DNPH1. Summary DNPH1 is a nucleotide-pool sanitizing hydrolase whose deletion selectively sensitizes homologous-recombination-deficient tumors to PARP inhibitors. To enable small-molecule validation of this synthetic-lethal target, AstraZeneca executed a fully integrated hit-finding program that combined high-throughput screening (HTS, 1.8 M compounds), DNA-encoded library (DEL, 7.1 billion compounds) affinity selection and fragment-based lead generation (FBLG). The campaign delivered four structurally distinct, non-nucleotide chemotypes—thiadiazine, imidazole, triazole and tetrahydro-isoquinoline (THIQ)—that were biophysically validated (IC₅₀ 2–24 µM; SPR Kd 2–9 µM) and structurally characterized by X-ray crystallography. Subsequent parallel optimization showed that only the thiadiazine series could be advanced to low-nM, cell-permeable inhibitors (e.g. compound 10: IC₅₀ 0.5 nM, cellular TE IC₅₀ 61 nM) and to potent PROTAC degraders (e.g. 11: DC₅₀ 28 nM). DEL-derived triazole ligands also furnished early PROTACs (e.g. 13) that achieved >90 % DNPH1 degradation before the more drug-like quinazoline/quinoline series became available. Imidazole and THIQ cores could not be driven below ~1 µM potency, illustrating the necessity of an acidic anchor for high-affinity binding and the penalty of stabilizing a folded bioactive conformation. Highlights First non-nucleotide ligands for DNPH1 discovered through a tri-platform approach (HTS + DEL + FBLG). Four validated chemotypes reveal divergent binding modes within a flexible, dimeric catalytic site. Thiadiazine → quinazoline core hop overcame permeability hurdles, yielding nM cell-active inhibitors and efficient PROTACs. DEL screen accelerated biology by enabling direct-to-biology PROTAC synthesis before lead-optimization completion. Structural and SAR data demonstrate that strong engagement of the phosphate-binding pocket (charged H-bond) is critical for sub-µM potency. Integrated screening maximized chemical coverage and mitigated single-technology failure (FBLG produced no confirmed hits). Conclusion By concurrently deploying HTS, DEL and FBLG, the team rapidly generated a diversified hit collection against the previously ligand-naïve target DNPH1. Crystal structures illuminated both opportunities and limitations: loop plasticity and the requirement for polar anchoring complicated optimization of neutral scaffolds, whereas acid-bearing thiadiazines were successfully morphed into quinazoline/quinoline analogues with single-digit nM enzymatic potency, robust cellular activity and efficient target degradation. A DEL-derived triazole further enabled early PROTAC proof-of-concept, underscoring the strategic value of exploiting the DNA-conjugation vector. Overall, the work delivers chemical tools that confirm DNPH1 as a druggable node in DNA-damage response pathways and exemplifies how an integrated discovery engine can de-risk and accelerate prosecution of challenging, novel targets within industrial timelines.

Yuichi Onda ,  Yurika Ochi ,  Toshihiro Araki ,  Miho Kageoka ,  Shuzo Takeda ,  Kazunori Yamada ,  Takehiko Ueda ,  Ken Ohno ,  Minoru Tanaka ,  Daiki Sakai ,  Miki Hasegawa ,  Yoshihito Tanaka  bioRxiv - Synthetic Biology DOI: 10.1101/2025.11.26.690606 Abstract Targeted protein degradation (TPD), including proteolysis targeting chimeras (PROTACs) and molecular glue degraders (MGDs), is a promising therapeutic approach. However, systematic discovery of such small molecules remains a major challenge. Here, we present PhenoDEL, a novel phenotypic DNA-encoded library (DEL) screening platform that integrates one-bead one-compound DEL (OBOC-DEL) with the Beacon® optofluidic system for high-throughput, single-cell analysis. By co-culturing individual OBOC-DEL beads and engineered reporter cells in nanoliter-scale chambers, PhenoDEL enables direct observation of compound-induced protein degradation at single-cell resolution. We demonstrate this approach by identifying compounds that induce degradation of FKBP12F36V-EGFP fusion proteins in PC-3 cells. The workflow allows precise linkage between compound identity and cellular phenotype via DNA barcoding and next-generation sequencing. PhenoDEL overcomes limitations of conventional screening methods, offering high sensitivity, spatial control, and scalability. This platform holds significant potential for mechanism-driven drug discovery, including identification of novel PROTACs and MGDs. Summary This preprint introduces PhenoDEL, a novel phenotypic screening platform that integrates One-Bead One-Compound DNA-Encoded Library (OBOC-DEL) technology with the Beacon® optofluidic system for high-throughput, single-cell analysis of targeted protein degradation (TPD). The platform enables direct observation of compound-induced protein degradation by co-culturing individual OBOC-DEL beads with engineered reporter cells in nanoliter-scale chambers (NanoPens). Upon UV-A irradiation, compounds are photoreleased from beads and diffuse to the cell, causing degradation of a FKBP12F36V-EGFP fusion protein. Beads associated with cells showing EGFP fluorescence loss are recovered, and their DNA barcodes are sequenced to identify active compounds. The authors optimized compound release kinetics (200 ms UV exposure), retention conditions (halting CO₂ flow increases concentration 5-fold), and imaging protocols. In a proof-of-concept screen, PhenoDEL successfully identified PROTAC molecules with >99% cell viability, demonstrating its capacity for mechanism-driven discovery of protein degraders including PROTACs and molecular glues. Highlights Single-cell resolution screening: PhenoDEL achieves 1:1 pairing of individual OBOC-DEL beads and cells in 0.75 nL NanoPen chambers, enabling direct linkage between compound identity and cellular phenotype via DNA barcoding. Real-time quality control: The platform excludes dead or damaged cells by monitoring EGFP fluorescence before and after compound release, reducing false positives and improving data reliability compared to pooled screening methods. Optimized compound delivery: UV-A irradiation (390 nm, 200 ms) cleaves photolabile linkers to achieve biologically relevant concentrations (10-90 µM) while stopping CO₂ flow enhances compound retention within chambers. High throughput: Up to four OptoSelect® chips (3,500 nanopens/chip) can run simultaneously, enabling >10,000 samples per run; scalable to 80,000 compounds using 20k-nanopen chips. Validated proof-of-concept: Engineered PC-3 cells expressing FKBP12F36V-EGFP showed robust degradation of the fusion protein within 6 hours of PROTAC FKBP Degrader-3 exposure, with minimal cytotoxicity and stable EGFP baseline (>5,000 fluorescence units). Versatile applicability: The system is compatible with various cell types (suspension, adherent, primary cells, organoids) and reporter systems, positioning it for personalized medicine and comprehensive functional genomics. Conclusion PhenoDEL represents a significant advancement in DNA-encoded library screening by overcoming limitations of conventional affinity-based and droplet-based methods. The integration of OBOC-DEL with Beacon's optofluidic technology enables high-resolution, activity-based screening at the single-cell level, providing spatial control, real-time phenotypic tracking, and direct genotype-phenotype correlation. The platform's ability to precisely modulate compound release, maintain cell viability, and automatically filter out low-quality data points establishes a robust framework for discovering novel PROTACs, molecular glue degraders, and other proximity-inducing molecules. With demonstrated scalability and compatibility across diverse biological models, PhenoDEL holds substantial potential for next-generation drug discovery, particularly in targeting previously undruggable proteins through event-driven pharmacology.

Li Zhou , Yong Ju , Zhijuan Cao , Sheng Cai , Jiayuan Su , Jianzhong Lu  Journal of Pharmaceutical Analysis DOI: 10.1016/j.jpha.2025.101498 Highlight Through screening 31 DNA-encoded chemical libraries, totaling 4.4 billion molecules, we identified a novel class of selective CDK2 inhibitors. The drug-likeness of C172 at the cellular level was evaluated, such as in vitro enzymatic and cellular assays, mechanistic studies on protein degradation, ADME characterization, single-dose pharmacokinetics in rats and metabolite identification.  

Yiwei Zhang, Yuqiu Lan, Rufeng Fan, Lei Feng, Guoliang Wang, Xinyuan Wu, Lulu Wen, Zhiqiang Duan, Yueyue Xia, Xudong Wang, Lingrui Zhang, Lu Zhou, Minjia Tan, Cangsong Liao, Xiaojie Lu  Journal of the American Chemical Society DOI: 10.1021/jacs.5c14634 Abstract DNA-encoded libraries (DELs) have emerged as an effective and efficient selection strategy for lead compound discovery in academia and industry over the past few decades. Despite recent advancements in this field, DEL remains limited by sensitive DNA-based constructs, particularly with low selection success rates resulting from the random selection of targets. Here, we describe a chemoenzymatic on-DNA reaction for DEL syntheses and develop a chemoproteomic-guided DEL selection platform. This platform, termed FF tags-biocatDEL, integrates DEL technology, chemoenzymatic synthesis, and fully functionalized (FF) chemical tags to match DELs with selection targets, even with limited information about ligandable hotspots. Using two diazirine-based FF indole probes, we comprehensively surveyed binding partners in cells and identified phosphoglycerate dehydrogenase (PHGDH) as a potential target for DEL selection. DEL01 and DEL02 were designed, synthesized, and selected against PHGDH, leading to the identification of a novel enzyme-active compound with an IC50 value of 2.5 μM. Our strategy, utilizing FF tags-biocatDEL, establishes a generalizable workflow for rapid target hunting and ligand discovery. It provides an effective method for precisely matching DELs with potential targets, demonstrating its significant potential as a complementary approach to drug discovery based on DELs. Summary This study presents a novel FF tags-biocatDEL platform that integrates chemoenzymatic synthesis, chemoproteomics, and DNA-encoded library (DEL) technology to overcome the low success rates of traditional DEL selection. The researchers developed a DNA-compatible decarboxylative aldol reaction using the PLP-dependent enzyme ApUstD to generate indole scaffolds bearing amine and carboxyl functional groups. Through chemoproteomic profiling with diazirine-based fully functionalized (FF) indole probes, they identified phosphoglycerate dehydrogenase (PHGDH) as a high-priority target from 2,208 enriched proteins. Two focused DELs were constructed: DEL01 (281,158 members via 2-cycle synthesis) and DEL02 (1.35 million members derived from a lactone fragment). Affinity selection against PHGDH yielded L5, a novel indole-based inhibitor with an IC₅₀ of 2.5 μM that acts via an allosteric mechanism. This strategy demonstrates that chemoproteomic guidance significantly enhances DEL selection efficiency and expands the chemical space for challenging targets. Highlights Innovative Chemoenzymatic Reaction: The first application of ApUstD on DNA substrates, achieving quantitative conversion (up to 100%) under mild aqueous conditions to generate complex indole scaffolds with γ-hydroxy-α-amino acid structures. Chemoproteomic-Guided Target Identification: Diazirine-based FF indole probes enabled unbiased profiling of 2,208 ligandable proteins, with PHGDH emerging as a clinically relevant target (ranked 111th) for cancer and neurodegenerative diseases. Potent Allosteric Inhibitor Discovery: L5, derived from a lactone byproduct (L3) scaffold, showed sub-micromolar potency (IC₅₀ = 2.5 μM) and allosteric inhibition independent of NAD⁺ concentration, representing a new chemotype for PHGDH. Scaffold Optimization via DEL Iteration: Initial hits (L1, L2) showed modest activity, but leveraging a side-product scaffold (L3) to build DEL02 (1.35M compounds) enabled a 20-fold activity improvement over the parent fragment. Technical Milestones: Successfully synthesized large-scale DELs using biocatalysis, validated target engagement via photo-crosslinking and thermal shift assays, and established a generalizable workflow combining fragment-based DELs with proteome-wide targeting data. Conclusion The FF tags-biocatDEL platform successfully bridges biocatalysis, chemoproteomics, and DEL technology to create a highly efficient, target-directed drug discovery workflow. By using chemoproteomic data to rationally select PHGDH and focused DELs to optimize a biocatalytically derived indole scaffold, the team discovered L5, a novel, compact PHGDH inhibitor with promising activity. This approach significantly outperforms random target selection and expands the accessible chemical space for traditionally challenging enzymes. While the platform currently leverages biocatalysis primarily for scaffold generation, future expansion to multiple DEL synthesis steps could further enhance diversity. Additionally, the affinity-based selection may identify non-functional binders that could be repurposed as PROTACs or other modalities. Overall, this strategy offers a robust complement to conventional DEL methods and holds substantial promise for accelerating lead discovery against emerging therapeutic targets.

Wenyi Zhang, Yuxing Wang, Rui Zhan, Runtong Qian, Qi Hu, Jing Huang bioRxiv - Biophysics DOI: 10.1101/2025.06.12.659183 Abstract DNA-encoded libraries (DELs) facilitate high-throughput screening of trillions of molecules against protein targets through split-pool synthesis and DNA tagging. Despite their potential, only a few DEL-derived compounds have advanced to clinical trials or reached the market. A better understanding of the defining characteristics of target proteins, particularly those with binding pockets suitable for DEL screening, is critical to improving success rates. However, existing approaches remain limited in assessing pocket flexibility and functional similarity. Here, we present ErePOC, a pocket representation model based on contrastive learning with ESM-2 embeddings to address these challenges. ErePOC captures both structural and functional features of binding pockets, enabling identification of shared characteristics among DEL targets. By integrating analyses of low-dimensional physicochemical properties and high-dimensional ErePOC embeddings, we provide a comprehensive view of DEL target space. With 98% precision in downstream classification tasks, ErePOC demonstrates high performance in pocket representation, which is then applied to predict human proteins suitable for DEL screening, with enrichment uncovered across 18 functional categories. This work establishes a new framework for enhancing DEL-based drug discovery through more effective target selection and pocket similarity analysis. Summary This study introduces ErePOC, a novel pocket representation model that employs contrastive learning with ESM-2 embeddings to decode the defining characteristics of protein binding pockets amenable to DNA-encoded library (DEL) screening. Despite DEL technology's capacity to screen trillions of compounds, clinical translation remains limited due to poor understanding of target druggability. The researchers analyzed 128 successful DEL targets and compared them to 326,416 general ligand pockets (BioLiP2) and 340 FDA-approved drug pockets, revealing that DEL pockets are uniquely larger (28.1 vs 16.1 residues), more hydrophobic, and enriched in specific amino acids (Met, Tyr, Trp, Phe, Leu). ErePOC was trained to map pockets to a 256-dimensional latent space aligned with ligand chemical similarity, achieving 98% precision in functional classification. Applied to 23,391 AlphaFold2-predicted human proteins, the model identified 2,739 DEL-compatible targets with pockets showing >0.8 cosine similarity to known DEL pockets. Enrichment analysis revealed 18 functional categories, particularly oxidoreductases, transferases, and multifunctional enzymes. In silico docking of 2.8 million virtual DEL compounds against 14 selected targets confirmed that ErePOC-enriched proteins exhibit significantly better predicted binding affinities than neutral controls. This work establishes a computational framework for rational DEL target selection beyond traditional structural similarity metrics. Highlights Distinct DEL Pocket Signature: DEL-binding pockets are 1.3× larger (3,301 ų volume), more hydrophobic (50.7% hydrophobic interactions vs 32.5% in natural pockets), and enriched in flexible aromatic/hydrophobic residues (Met, Tyr, Trp, Phe) compared to regular ligand and FDA-approved drug pockets. ErePOC Model Innovation: A contrastive learning framework that aligns pocket representations with ligand Morgan fingerprints via KL divergence loss, generating function-aware embeddings that capture physicochemical and evolutionary features beyond 3D geometry, robust to pocket flexibility. Robust Zero-Shot Performance: Achieves superior classification of 7 ligand-binding pocket types (~43,000 pockets) with 98.5–98.9% accuracy; maintains strong performance even for pocket classes excluded from training, demonstrating powerful generalization. Large-Scale Human Proteome Screening: Identified 2,739 unique human proteins with DEL-compatible pockets from AlphaFold2 structures, with significant enrichment in transferases (17.9%), hydrolases (11.6%), and oxidoreductases (9.4%), plus novel classes like RNA-binding proteins and chromatin regulators. Experimental Validation via Docking: In silico screening of 2.8M DEL-like molecules against ErePOC-selected targets showed statistically significant better binding affinity (mean Z-score –2.18 vs –1.07) and higher enrichment for DEL-enriched vs DEL-neutral protein families. Case Study Insights: The regulatory protein MAT2B exhibits higher DEL compatibility (cosine similarity 0.93, docking –8.8 kcal/mol) than its catalytic paralog MAT2A (0.66, –5.3 kcal/mol), demonstrating ErePOC's ability to resolve subtle family-level differences in druggability. Conclusion ErePOC provides a transformative approach to DEL target selection by learning high-dimensional, function-aware representations of binding pockets that transcend traditional structural alignment limitations. The model successfully deciphers a unique DEL pocket pattern—characterized by larger size, enhanced hydrophobicity, and specific amino acid biases—and leverages this to predict over 2,700 human proteins likely amenable to DEL screening across 18 enriched functional categories. By capturing physicochemical relationships rather than relying solely on geometric similarity, ErePOC addresses the critical challenge of pocket flexibility and low structural overlap among functionally related sites. The significant enrichment of oxidoreductases, transferases, and multifunctional enzymes validates known DEL success stories while expanding the targetable space to include chromatin regulators and RNA-binding proteins. In silico validation confirms that ErePOC-selected targets bind DEL-like molecules more favorably, supporting its practical utility. This framework not only enhances DEL efficiency but also offers broad applicability for virtual screening, molecule generation, and protein design, particularly when integrated with advanced structure prediction tools like AlphaFold3.

Mengke You, Hong Liu, Chunpu Li JACS Au  DOI: 10.1021/jacsau.5c01134 Abstract Covalent modification of therapeutic targets has emerged as a powerful platform for creating clinical drugs and chemical probes. Covalent drugs have evolved from serendipitous discoveries to rationally designed therapeutics, driven by advances in electrophile-first screening technologies. This perspective takes stock of alternative technologies currently available in laboratories and industry that collectively enable targeted covalent inhibitor development across historically “undruggable” targets. We highlight five such technologies: activity-based protein profiling (ABPP), provides functional proteomic mapping to identify ligandable residues; covalent tethering, exploits dynamic chemistry to capture transient pockets; covalent DNA-encoded libraries, leverages trillion-member libraries for multiresidue targeting; phage/mRNA display, which facilitates evolution of covalent macrocyclic peptides; and sulfur(VI) fluoride exchange (SuFEx), engages residues beyond cysteine. Integration of these approaches with chemoproteomics and artificial intelligence accelerates the discovery of covalent inhibitors with enhanced selectivity and reduced off-target risks. This technological convergence establishes a new paradigm for precision covalent therapeutics, offering innovative solutions to overcome drug resistance and target challenging protein interfaces.

Hossam Nada ,  Shaoren Yuan ,  Farida El Gaamouch ,  Sungwoo Cho ,  Katarzyna Kuncewicz ,  Laura Calvo-Barreiro ,  Moustafa T. Gabr European Journal of Medicinal Chemistry  DOI: 10.1016/j.ejmech.2025.118358 Abstract Triggering receptor expressed on myeloid cells 2 (TREM2) is a key regulator of microglial function, and its loss-of-function variants are linked to Alzheimer’s disease (AD) and neurodegenerative disorders. While TREM2 activation is a promising therapeutic strategy, no small molecule agonists acting via direct TREM2 binding have been reported to date. Here, we describe the discovery of first-in-class, direct small molecule TREM2 agonists identified through DNA-encoded library (DEL) screening. The DEL hit (4a) demonstrated TREM2 binding affinity, as validated by three biophysical screening platforms (TRIC, MST, and SPR), induced Syk phosphorylation, luciferase assay and enhanced microglial phagocytosis. Pre-liminary optimization yielded 4i, which maintained TREM2 engagement with improved selectivity over TREM1 and no cytotoxicity. Molecular dynamics simulations predicted that 4a stabilizes a transient binding pocket on TREM2, indicating the possibility of a novel mechanism for receptor activation. These findings provide the first proof-of-concept for direct pharmacological TREM2 agonism, offering a foundation for developing therapeutics against AD and related disorders.

Roy Aerts, Joris Tavernier, Alan Kerstjens, Mazen Ahmad, Jose Carlos Gómez-Tamayo, Gary Tresadern, Hans De Winter Journal of Cheminformatics DOI: https://doi.org/10.1186/s13321-025-01109-x Abstract Peptide drug development is currently receiving due attention as a modality between small and large molecules. Therapeutic peptides represent an opportunity to achieve high potency, selectivity, and reach intracellular targets. A new era in the development of therapeutic peptides emerged with the arrival of cyclic peptides which avoid the limitations of parenteral administration via achieving sufficient oral bioavailability. However, improving the membrane permeability of cyclic peptides remains one of the principal bottlenecks. Here, we introduce a deep learning regression model of cyclic peptide membrane permeability based on publicly available data. The model starts with a chemical structure and goes beyond the limited vocabulary language models to generalize to monomers beyond the ones in the training dataset. Moreover, we introduce an efficient estimator2generative wrapper to enable using the model in direct molecular optimization of membrane permeability via chemical modification. We name our application C2PO (Cyclic Peptide Permeability Optimizer). Lastly, we demonstrate how a molecule correction tool can be used to limit the presence of unfamiliar chemistry in the generated molecules. Summary This study presents C2PO (Cyclic Peptide Permeability Optimizer), a novel machine learning-driven application that improves the membrane permeability of cyclic peptides through chemical structure modification. The core of C2PO consists of a Graph Transformer deep learning model trained on the CycPeptMPDB dataset (7,451 permeability measurements), achieving state-of-the-art performance (R² = 0.61, Pearson r = 0.78, MAE = 0.37 on test set). Unlike conventional generative models, C2PO employs an estimator2generative approach, using gradient-based optimization based on the HotFlip algorithm to suggest structural modifications. The framework operates in two stages: first, it generates permeability-optimized peptide analogs by mutating side chains while preserving the macrocycle backbone; second, it automatically corrects chemically invalid structures using a dictionary-based correction tool referencing ChEMBL31. A case study on 700 low-permeability cyclic peptides demonstrated that 76.86% of optimization campaigns successfully produced at least one offspring with improved permeability (logPapp > -6.0), with 42.05% of all 13,043 generated molecules crossing this threshold. The system allows flexible user control over modification scope, elemental composition, and optimization parameters, making it a practical tool for medicinal chemists to generate ideas for improving peptide drug candidates. Highlights First-in-class application converting a machine learning model into a generative optimizer specifically for cyclic peptide permeability improvement Estimator2generative paradigm that decouples property estimation from structure generation, enabling broader chemical space exploration beyond training vocabulary State-of-the-art Graph Transformer model (based on GRAPHGPS framework) trained on comprehensive CycPeptMPDB dataset with robust cross-validation performance Automated chemistry correction workflow using a dictionary-based tool to validate and fix chemically unrealistic structures post-optimization, preserving 78% of successful optimizations Demonstrated effectiveness in a large-scale case study: 76.86% success rate for campaigns and 42.05% of offspring molecules achieving high permeability High flexibility allowing user-defined constraints on backbone protection, elemental modifications, molecular size changes (±5 atoms), and optimization parameters Beyond-vocabulary generalization capability to handle monomers not present in training data, overcoming limitations of language model-based approaches Conclusion Generally, cyclic peptides lack adequate membrane permeability to be developed into medicines. We propose C2PO (Cyclic Peptide Permeability Optimizer), an application that improves permeability by modifying the chemical structure of a given cyclic peptide. C2PO is ML-driven, trained on the experimental CycPeptMPDB dataset, and can be categorized in the estimator2generative optimization paradigm. However, ML-based applications that output chemical structures have the tendency of occasionally proposing strange chemistry, attributable to the loss of chemical knowledge, although it is generally considered to be implicitly learned. Therefore, we opted for checking and correcting the outcomes of C2PO using a chemistry library-based autocorrection application in a subsequent step. This contribution provides insights into what one can expect when applying these two applications. Seven hundred permeability optimization campaigns were launched where only peptide side chains were allowed to be modified. In general, we observed optimization for many campaigns, meaning that bad permeability starting points were optimized to structures with estimated permeability above the threshold of -6.0 logPapp. In the chemical correctness check step, we identified that a substantial portion (22.9%) of output structures needed correction. The autocorrection tool modified these, and we tracked how optimized permeability altered upon chemical correction. Various scenarios occurred, but the most important was that for many campaigns, the second step did not counteract the initial permeability optimization. We discussed in detail how to properly use our model and workflow, noting its flexibility for user customization. We focused on providing insights into basic capabilities rather than pursuing optimal performance, while informing about ways to improve both permeability optimization and molecular autocorrection. Finally, we hope to raise general interest in adopting estimator2generative optimizer strategies for chemical problems and deploying chemistry-library-driven applications for post-correcting ML-generated structures.

Ji Young Ryu ,  Dohui Ku ,  Hayeong Shin ,  Yoojin Park ,  Hokyung Kim ,  Jihoon Lee ,  Gil Tae Hwang ,  Minsoo Song ,  Ki Tae Kim Organic Letters DOI: 10.1021/acs.orglett.5c04328 Abstract The development of new on-DNA synthetic methods is crucial for efficient drug discovery using DNA-encoded libraries (DELs). Herein, we present a novel on-DNA approach based on hydrosulfonation of alkynylcarbonyl compounds to access β-sulfonyl-α,β-unsaturated carbonyl scaffolds bearing vinyl and γ-carbonyl sulfone skeletons. The reaction proceeds via a simple mix of alkynylcarbonyl compound and sodium sulfinate at 80 °C, showing moderate to good conversions, Z-geometry preference, and good DNA compatibility, supporting its utility in DEL construction.

Yagong Wang ,  Kehan Zhou ,  Haiqian Zhu ,  Huimin Sun ,  Hongli Cao ,  Lijun Xue ,  Kexin Yang ,  Yun Jin Hu Organic Letters DOI: 10.1021/acs.orglett.5c04222 Abstract We present a modular on-DNA bridging strategy that addresses key reactivity challenges in DNA-encoded library synthesis. By generation of a reactive carbamodithioic acid intermediate from secondary amines, this method enables efficient coupling with epoxides to produce diverse 2-hydroxy dithiocarbamate scaffolds. Gentle and DNA-compatible, it offers a versatile platform for rapid, structurally diverse DEL construction with significant drug discovery potential.

Yi Gan, Yumei Zeng, Haojie Guan, Wenjun Li, Alex Shaginian, Jin Li, Sen Gao, Guansai Liu Bioconjugate Chemistry https://doi.org/10.1021/acs.bioconjchem.5c00455 Abstract Bicylic peptides, with two cyclic substructures, have emerged as a powerful tool for modulating challenging targets such as protein−protein interactions. Meanwhile, DNA-encoded library technology (DELT) provides a powerful platform for hit discovery. The unity of both fields has the potential to identify potent bicyclic ligands for the targets of interest. Therefore, there is a high demand to develop an efficient way to construct bicyclic peptide libraries. Herein, we describe a novel and efficient approach to the synthesis of DNA-encoded bicyclic peptides via a cysteine-promoted cyclization and amide condensation reaction. This strategy proceeds smoothly under mild conditions and can generate a wide range of bicyclic peptides with various peptide sequences and ring sizes in good conversion. The method employs a trifunctional cross-linker, methyl 3,5-bis-(bromomethyl)benzoate, to enable sequential thioether and amide bond formation, overcoming stability issues associated with traditional protecting groups. The chemistry is compatible with diverse functional groups, N-methylation, and both short and long DNA strands, demonstrating its suitability for constructing large-scale DNA-encoded bicyclic peptide libraries. Summary This report presents a robust DNA-compatible methodology for synthesizing bicyclic peptides suitable for DNA-encoded library technology (DELT). The strategy addresses a critical limitation in the field: the instability of StBu-protected cysteine under Fmoc-deprotection conditions, which generates byproducts and compromises library quality. The authors developed a two-step macrocyclization approach using methyl 3,5-bis-(bromomethyl)benzoate as a trifunctional cross-linker. The first cyclization forms a dithioether linkage via reaction with cysteine thiols, while the second cyclization creates an amide bond through ester hydrolysis and intramolecular condensation. This method proceeds under mild aqueous conditions and demonstrates broad substrate scope, enabling synthesis of bicyclic peptides with ring sizes ranging from [2+2] to [7+8] amino acids. The chemistry tolerates various functional groups (indole, guanidine, phenol, thioether, etc.) and backbone N-methylation. Validation studies on 543 di-, tri-, and tetrapeptide building blocks showed 420 achieved ≥50% conversion, confirming feasibility for library production. The method was successfully applied to both short DNA (7 bp) and long DNA (72 bp) with enzymatic ligation, showing no significant DNA damage. Off-DNA synthesis confirmed the structural integrity of the bicyclic products. This approach enables access to billions of diverse bicyclic peptides, bridging the therapeutic gap between small molecules and biologics. Highlights Wide substrate scope: Successfully synthesized bicyclic peptides with diverse amino acid sequences and ring sizes from [2+2] to [7+8] High efficiency: Achieved conversions of 45-88% across various bicyclic peptide structures Mild reaction conditions: Compatible with DNA-encoded library synthesis requirements Two-step macrocyclization: Novel use of trifunctional cross-linker enabling sequential thioether and amide bond formation Functional group tolerance: Compatible with indole, guanidine, phenol, thioether, alcohol, and N-methylated residues DEL compatibility: Validated on long-chain DNA (72 bp) with successful enzymatic ligation and no detectable DNA damage Scalability potential: Demonstrated feasibility for constructing libraries exceeding 1.6 billion compounds using validated di/tri/tetrapeptide building blocks Conclusion In summary, we have developed a novel and efficient methodology to synthesize DNA-encoded bicyclic peptides via two independent macrocyclization reactions. The employment of a trifunctional cross-linker, methyl 3,5-bis-(bromomethyl)benzoate, enabled both cysteine-promoted cyclization and amide cyclization chemistries. A large number of bicyclic peptides with a wide range of structurally diverse backbones and side-chains were constructed with high efficiency. The commercial availability of building blocks (amino acids and peptides) as well as the broad scope exploration demonstrated the feasibility of our protocol for the preparation of a structurally diverse library of bicyclic peptides. Efforts to transform this robust and efficient chemistry to bicyclic peptide DEL construction will be continued in due course.

Maxwell Sigal ,  Markus Egner ,  Chikako Okada ,  Daniel Merk ,  Toru Sengoku ,  Takayuki Katoh ,  Hiroaki Suga Journal of the American Chemical Society  DOI: 10.1021/jacs.5c13803 Abstract α,α-Disubstituted α-amino acids (dαAAs) are important building blocks for peptidomimetics as they are strong inducers of helicity and protect against proteolytic degradation. However, de novo discovery of dαAA-containing peptides with genetically encoded libraries is limited due to their poor incorporation efficiency. Here, we report the optimized ribosomal incorporation of multiple achiral dαAAs into peptide libraries and their application to high-throughput (>1012 members) affinity selection against the nuclear receptor peroxisome proliferator-activated receptor-gamma (PPARγ). This dαAA-based screening methodology discovered potent linear and macrocyclic α-helical peptides with low-to-sub nanomolar binding affinities. Hit peptides were proteolytically stable in serum and cell permeable, allowing for in cellulo antagonism of PPARγ. X-ray crystallography revealed that dαAA-containing peptides bound at the α-helical protein–protein interaction (PPI) interface via an α-helical conformation. This work validates the potential of a dαAA-based, α-helical discovery platform, providing access to new chemical and conformational space to de novo identify novel α-helical peptidomimetics.

Yagong Wang ,  Hangke Ma ,  Ivan Hu ,  Jin Liang ,  Bingxin Chen ,  Lijun Xue ,  Kexin Yang ,  Yun Jin Hu Organic Letters  DOI: 10.1021/acs.orglett.5c04043 Abstract We present a metal-free photochemical strategy for the synthesis of on-DNA β-hydroxy sulfides, which are essential pharmacophores with limited access via existing DNA-encoded library (DEL) methodologies. This innovative hydrothiolation features a broad substrate scope with both thiols and olefins. Its operational simplicity and excellent DNA compatibility enable the efficient construction of diverse DELs, significantly expanding the accessible chemical space for drug discovery and providing a new tool for identifying novel therapeutics. Summary This study introduces a novel metal-free, photoinduced method for synthesizing β-hydroxy sulfides on DNA. The approach leverages visible light to drive the reaction, avoiding the use of metal catalysts that can be harmful to DNA. The protocol is characterized by its operational simplicity, broad substrate scope, and excellent compatibility with DNA, making it highly suitable for constructing diverse DNA-encoded libraries (DELs). This method significantly expands the chemical space accessible for drug discovery and offers a powerful tool for identifying new therapeutic agents. Highlights Metal-Free and Photoinduced: The method is metal-free and utilizes visible light, avoiding conditions harmful to DNA. Broad Substrate Scope: The reaction tolerates a wide range of thiols and olefins, including both aromatic and aliphatic substrates. Operational Simplicity: The protocol is straightforward, requiring only readily available starting materials and mild reaction conditions. DNA Compatibility: The method is highly compatible with DNA, enabling efficient construction of DELs without significant DNA damage. Expands Chemical Space: This approach significantly broadens the chemical space accessible for DEL-based drug discovery. Conclusion The development of a metal-free, photochemical β-hydrothiolation for the on-DNA synthesis of β-hydroxy sulfides represents a significant advancement in the field of DNA-encoded library synthesis. This operationally simple protocol demonstrates excellent functional group tolerance and DNA compatibility, enabling the efficient construction of a diverse DNA-encoded library from readily available thiols and disulfides. The method provides a powerful and versatile tool that significantly expands the accessible chemical space for DEL-based drug discovery, offering a new avenue for identifying novel therapeutic agents.

Wei Zhou, Yan Wang, Shuning Zhang, Chengwei Zhang, Jiacheng Pang, Shaoneng Hou, Jie Li, Ying Yao, An Su, Peixiang Ma, Hongtao Xu, Wei Hou Chemical Science  DOI: 10.1039/d5sc05512a Abstract Chemical reactions compatible with multiple functionalities are essential for rapid, programmable, and automatable synthesis of functional molecules. However, achieving such reactivity poses significant challenges. Here, we developed a novel multi-orthogonal C(sp 2 )-Se bond formation reaction between benzoselenazolones and boronic acids via Ag(I)catalyzed selective selenium(II)-nitrogen exchange. This chemistry is compatible with diverse functionalities, enabling sequential and programmable synthesis. Moreover, it features modular, high-yielding (485 examples, with yields or conversions exceeding 70% in 95% cases), and switchable reaction systems under mild conditions. Its practical utility was exemplified through late-stage functionalization of natural products, peptides modification and ligation, diversified synthesis, sequential click chemistry, protecting group-free syntheses of sequence-defined oligoselenide (nonamer), on-plate nanomole-scale parallel synthesis (200 nmol, 412 selenides), and DNA-encoded library (DEL) synthesis (10 nmol, 92 examples). Notably, a target-based screening identified SA-16 as a potent CAXII inhibitor with an IC 50 value of 72 nM. Furthermore, a machine learning-based model (SeNEx-ML) was established for reaction yield prediction, achieving 80% accuracy in binary classification and 70% balanced accuracy in ternary classification. These results demonstrated that this chemistry serves as a powerful tool to bridge the selenium chemical space with the existing chemical world, offering transformative potential across multidisciplinary fields. Summary This article presents a novel, highly selective, and AI-predictable selenium(II)-nitrogen exchange (SeNEx) chemistry between benzoselenazolones and boronic acids. The Ag(I)-catalyzed reaction forms C(sp²)-Se bonds and is compatible with a wide range of functionalities, making it suitable for programmable and parallel synthesis. The chemistry is modular, high-yielding, and operates under mild conditions. It has been successfully applied in various practical syntheses, including late-stage functionalization of natural products, peptide modifications, and DNA-encoded library synthesis. A machine learning model (SeNEx-ML) was also developed to predict reaction yields with high accuracy. Highlights Development of a novel Se-N exchange chemistry between benzoselenazolones and boronic acids. The reaction is highly selective, modular, and high-yielding, with 95% of 485 examples achieving yields or conversions exceeding 70%. Compatibility with diverse functionalities and orthogonal to other top reactions in medicinal chemistry. Successful application in late-stage functionalization, peptide modification, and DNA-encoded library synthesis. Establishment of a machine learning model (SeNEx-ML) for reaction yield prediction with high accuracy. Conclusion In summary, we have successfully designed and developed an unprecedented highly selective and multi-orthogonal SeNEx chemistry between benzoselenazolones and boronic acids. This chemistry features modular, predictable, robust, and high-yielding characteristics, performed under mild and switchable reaction conditions. It demonstrates exceptional chemo-selectivity and functional group tolerance, enabling orthogonal synthesis with other established reactions. The practical applications in late-stage modification, peptide ligation, and DNA-encoded library synthesis highlight its potential in multidisciplinary fields. The establishment of the SeNEx-ML model further enhances its utility by predicting reaction outcomes accurately. This chemistry serves as a powerful tool to bridge the selenium chemical space with the existing chemical world, offering transformative potential in synthetic chemistry, material science, chemical biology, and medical chemistry.

Edith van der Nol ,  Nils Alexander Haupt ,  Qing Qing Gao ,  Benthe A. M. Smit ,  Martin Andre Hoffmann ,  Martin Engler-Lukajewski ,  Marcus Ludwig ,  Sean McKenna ,  J. Miguel Mata ,  Olivier J. M. Béquignon ,  Gerard van Westen ,  Tiemen J. Wendel ,  Sylvie M. Noordermeer ,  Sebastian Böcker ,  Sebastian Pomplun Nature Communication DOI: 10.1038/s41467-025-65282-1   Abstract Affinity-selection platforms are powerful tools in early drug discovery, but current technologies – most notably DNA-encoded libraries (DELs) – are limited by synthesis complexity and incompatibility with nucleic acid-binding targets. We present a barcode-free self-encoded library (SEL) platform that enables direct screening of over half a million small molecules in a single experiment. SELs combine tandem mass spectrometry with custom software for automated structure annotation, eliminating the need for external tags for the identification of screening hits. We develop efficient, high-diversity synthesis protocols for a broad range of chemical scaffolds and benchmark the platform in affinity selections against carbonic anhydrase IX, identifying multiple nanomolar binders. We further apply SELs to flap endonuclease 1 (FEN1) – a disease related DNA-processing enzyme inaccessible to DELs – and discover potent inhibitors. Taken together, screening barcode-free libraries of this scale all at once represents an important development, enables access to novel target classes, and promises substantial impact on both academic and industrial early drug discovery. Summary This article introduces a barcode-free self-encoded library (SEL) platform for affinity selection-based hit discovery in early drug discovery. The SEL platform leverages tandem mass spectrometry and custom software for automated structure annotation, eliminating the need for external tags. The platform enables the screening of large libraries with diverse scaffolds, identifying high-affinity binders for targets like carbonic anhydrase IX and inhibitors for flap endonuclease 1 (FEN1). This approach overcomes limitations of DNA-encoded libraries (DELs) by avoiding synthesis complexity and incompatibility with nucleic acid-binding targets, offering a streamlined workflow for rapid hit discovery. Highlights 1. A barcode-free self-encoded library (SEL) platform enables screening of over half a million small molecules in a single experiment. 2. SELs combine tandem mass spectrometry with custom software for automated structure annotation, eliminating the need for external tags. 3. Efficient, high-diversity synthesis protocols are developed for a broad range of chemical scaffolds. 4. The platform identifies multiple nanomolar binders for carbonic anhydrase IX and potent inhibitors for flap endonuclease 1 (FEN1). 5. SELs offer advantages in synthesis complexity and scope compared to DNA-encoded libraries (DELs), expanding the accessible chemical space for drug discovery. Conclusion The SEL platform represents a significant advancement in early drug discovery by enabling the screening of large, diverse libraries without the need for external tags. This barcode-free approach overcomes limitations associated with DNA-encoded libraries, such as synthesis complexity and incompatibility with nucleic acid-binding targets. The successful identification of high-affinity binders and inhibitors for challenging targets like carbonic anhydrase IX and flap endonuclease 1 demonstrates the platform's potential for rapid hit discovery. The streamlined synthesis and automated decoding capabilities of SELs make this technology accessible to a broader range of researchers, promising substantial impact on both academic and industrial drug discovery efforts. Future work will focus on further optimizing the platform for even larger libraries and exploring its application to additional challenging targets.

Alice R. Wong ACS Medicinal Chemistry Letters DOI: 10.1021/acsmedchemlett.5c00356 Abstract DNA-encoded libraries (DELs) have revolutionized hit identification in drug discovery by offering an accessible, versatile, and cost-effective alternative to traditional high-throughput screening (HTS). This perspective analyzes the results of recent DEL technology (DELT) screens (2020−2025) to enable medicinal chemistry programs, focusing on case studies where active series were generated from DEL and distills key learnings and design principles for productive library construction. A notable trend is the productivity of 2-cycle DELs, which, despite their smaller size, consistently yield hits and have superior physicochemical properties compared to 3-cycle DELs. The criteria for inclusion are where DEL provided a medicinal chemistry series, defined by off-DNA hit resynthesis, profiling in relevant assay(s), and follow-up SAR optimization. Summary This article provides an in-depth analysis of the design and application of DNA-encoded libraries (DELs) in medicinal chemistry. It examines recent case studies (2020−2025) where DELs have been successfully used to generate active series for drug discovery. Key learnings include the effectiveness of 2-cycle DELs in yielding hits with better physicochemical properties compared to 3-cycle DELs. The article also explores various DEL designs, including linear, branched, and heterocycle-formation designs, and highlights the importance of physicochemical properties in hit identification. The study concludes that while there is no clear correlation between DEL size and productivity, 2-cycle libraries have shown significant promise in generating high-quality hits. Highlights DNA-encoded libraries (DELs) offer a powerful alternative to traditional high-throughput screening (HTS) for hit identification. Recent case studies (2020−2025) demonstrate the effectiveness of DELs in generating active series for medicinal chemistry. 2-cycle DELs, despite their smaller size, consistently yield hits with superior physicochemical properties compared to 3-cycle DELs. Key physicochemical properties, such as molecular weight (MW), topological polar surface area (TPSA), and hydrogen bond donors (HBD), are critical in library design. The article emphasizes the importance of well-established on-DNA chemistry and standard building block classes in generating diverse and high-quality DELs. Conclusion The analysis of recent DEL technology screens highlights the potential of 2-cycle DELs in generating high-quality hits with desirable physicochemical properties. While there is no clear correlation between library size and productivity, 2-cycle libraries have shown significant promise. The study underscores the importance of physicochemical properties in hit identification and the effectiveness of well-established on-DNA chemistry and standard building block classes in library design. Future DEL designs should continue to leverage these principles to accelerate drug discovery.

Susmith Mukund Structural Dynamics DOI: 10.1063/4.0001022 Abstract The methyl thioadenosine phosphorylase (MTAP) gene which is proximal to the CDK2N2A tumor suppressor gene on chromosome locus p23q is frequently deleted in ∼15% of cancer cells. This results in the accumulation of methylthioadenosine (MTA), which competes with S-adenosyl methionine (SAM), the methyl donor for the essential enzyme, protein arginine methyltransferase 5 (PRMT5). PRMT5 is thereby put in a hypomorphic state in these MTAP-deleted cancer cells, presenting an opportunity for its further inhibition with MTA-cooperative inhibitors. DNA- encoded library screening produced hits that cooperatively bound PRMT5:MEP50 and MTA. Optimization of these compounds and structural enablement through both crystallography and cryo-electron microscopy led to the discovery of AMG 193. Crystal structure shows AMG 193 occupying the peptide binding site of the PRMT5 catalytic domain, where the R3 of the peptide substrate binds during the catalytic cycle, and in the vicinity of the co-inhibitor MTA (see figure, PDB id: 9C10). The tricyclic dihydrofuro- naphthyridine warhead mimics the R3 of the substrate with similar interactions, its amino-heterocyclic moiety forming salt bridge with E444 acid and a H-bond with the backbone carbonyl of E435, the furan oxygen in a H-bond with K333. The amino-heterocycle is in a displaced π:π stacking between W579 and F327, as well as van der Waals contact with residue Glu435, and MTA, a key feature of the MTA cooperative nature of the inhibitor. AMG 193 is further stabilized by a H-bond between its amide carbonyl, which is perpendicular to the tricyclic core, and the backbone amide of F580. The morpholine substituent adopts an axial conformation relative to tricyclic warhead. The terminal trifluoro-phenyl is packed in the very hydrophobic distal end of the substrate pocket, F580 in a π:π stack and Y304 in an edge-to-face stack with the phenyl ring of the ligand. The structure further explains why AMG 193 is MTA-cooperative and not synergistic with SAM.

Orville Pemberton, Amanda M Nevins, Thomas E Frederick, Emily Nicholl, Myron Srikumaran, Jun Chen, Alla Korepanova, Vincent Stoll, Andrew Petros, Sujatha Gopalakrishnan, Justin Dietrich, Liliam Rios Cordero, David J Hardee, Teresa I Ng, Chaohong Sun  Structural Dynamics DOI: 10.1063/4.0000814 Abstract The COVID-19 pandemic caused by SARS-CoV-2 has devastated global health, revealing an urgent need for novel therapeutics. The papain-like protease (PLpro) is one of two proteases encoded by SARS-CoV-2, representing an attractive drug target due to its dual roles in viral replication and host immune suppression. We employed a DNA-encoded library (DEL) screen to reveal starting points for our PLpro hit discovery campaign. These efforts led to the identification of compound 1, a diarylmethanol-containing substructure with a unique binding mode that induces PLpro dimerization. Compound 1 demonstrates potent activities in both a biochemical ubiquitin-rhodamine and antiviral HeLa-ACE2 cell assays. An X- ray co-crystal structure of compound 1 bound to PLpro solved to 2.0 Å showed that two molecules of compound 1 glues two monomers of PLpro together via binding to the BL2 groove of one monomer and the Ubl/thumb domain of the other. Several molecular interactions were observed between compound 1 and PLpro including hydrophobic interactions and several hydrogen-bonds across both monomers. The molecular glue-like properties of compound 1 on PLpro were further validated in solution with analytical SEC and protein-detect 2D-NMR. Subsequent rounds of SAR led to compound 2, which has comparable biochemical and antiviral activities and demonstrated the same dimerization mechanism of PLpro as seen in a 1.8 Å X-ray co-crystal structure. In summary, we have identified a new series of PLpro inhibitors with a novel mechanism of SARS-CoV-2 inhibition, providing a promising start for the discovery of antivirals for treating COVID-19.

Katarzyna B. Handing, Mu-Sen Liu, Douglas A. Whittington, Sining Sun, Rebecca Salerno, William D. Mallender, Jon Come, Scott Throner, Andrew Maynard, Patrick McCarren, John P. Maxwell, Serge Gueroussov, Kiera Vassallo, Yingnan Chen, Jannik N. Andersen, Wenhai Zhang Molecular Cancer Therapeutics DOI: 10.1158/1535-7163.targ-25-c104 Abstract MGAT1 is an N-glycosyltransferase essential for the synthesis of N-glycans. Cell surface glycans serve as immune checkpoints, playing a key role in cancer immune evasion. Knockout (KO) of MGAT1 enhances immune recognition and promotes T-cell–mediated killing, with enzymatic activity being necessary for this phenotype. These findings position MGAT1’s catalytic function as an attractive target for cancer therapy. In this study, we report the discovery of novel MGAT1 binders and inhibitors with sub-micromolar potency. Compounds were identified through two independent screening approaches: a UDP-GloTM-based high-throughput screen (HTS) measuring inhibition of MGAT1 enzymatic activity, and a DNA-encoded library (DEL) screen selecting for molecules that reproducibly bind to MGAT1. High-resolution crystal structures reveal detailed interactions between MGAT1 and the compounds, clearly identifying a binding site distinct from the active site. Surface plasmon resonance (SPR) competition assays further demonstrate that these inhibitors bind noncompetitively with respect to the endogenous product, UDP. Together, these results validate allosteric inhibition of MGAT1 as a novel and tractable strategy for impeding MGAT1 activity. Additionally, our findings lay the foundation for future structure-based optimization of MGAT1 inhibitors with potential application in cancer immunotherapy.

Johannes Bingold, Erik Mafenbayer, Wibke Langenkamp, Lisa Liang, Chun Zhang, Malte Mildner, Julia Isabel Bahner, Mohamed Akmal Marzouk, Bettina Böttcher, Ann-Christin Pöppler, Ralf Weberskirch, Andreas Brunschweiger Chemical Science  DOI: 10.1039/d5sc06782k Abstract Chemical diversification of DNA-conjugated substrates is key in DNA-encoded library (DEL) synthesis and other nucleic acid-based technologies. One major challenge to the translation of synthesis methods to DNA-tagged substrates is the lack of solubility of the highly charged DNA oligomer in most organic solvents. A neutral acrylate block copolymer, devoid of any canonical nucleic acid-binding structure, tightly interacted with DNA oligonucleotides in their ammonium form, and solubilized them in nonpolar solvents such as dichloromethane, chloroform and toluene. The ternary DNA–copolymer–ammonium salt interactions led to the formation of aggregates in organic solvents whose size correlated with DNA oligomer length. This method for DNA solubilization was successfully applied to diversify DNA-tagged starting materials by three isocyanide multicomponent reactions (IMCR) with broad scope and excellent yields. The copolymer does not require tailored DNA conjugates and solubilized DNA oligomers of up to 80 nucleotides length. It will likely broaden the toolbox of DEL-compatible synthesis methods well beyond IMCR chemistry and it has application potential in other nucleic acid-based technologies that require a broadened solvent scope for nucleic acid conjugate synthesis. Summary This article presents a novel method for solubilizing DNA oligomers in organic solvents using a neutral acrylate block copolymer. The method leverages ternary interactions between the DNA, the copolymer, and ammonium salts to form aggregates in nonpolar solvents. This approach enables the use of DNA-tagged substrates in a variety of chemical reactions, including isocyanide multicomponent reactions (IMCR), which are typically challenging due to the poor solubility of DNA in organic solvents. The study demonstrates the successful application of this method to diversify DNA-tagged starting materials with excellent yields and broad scope, potentially expanding the range of DEL-compatible synthesis methods. Highlights 1. A neutral acrylate block copolymer solubilizes DNA oligomers in nonpolar organic solvents. 2. Ternary interactions between DNA, copolymer, and ammonium salts form aggregates in organic solvents. 3. The method enables DNA-tagged substrates to undergo isocyanide multicomponent reactions (IMCR) with high yields. 4. DNA oligomers of up to 80 nucleotides length can be solubilized using this approach. 5. The copolymer system is compatible with downstream DEL operations such as enzymatic DNA tag ligation and barcode amplification. Conclusion The study introduces a conceptually novel approach to DNA solubilization in organic solvents using a neutral acrylate block copolymer. This method, termed CECOS (copolymer-mediated encoded chemistry in organic solvents), facilitates the translation of three isocyanide multicomponent reactions to DNA-tagged substrates with excellent yields and broad substrate scope. The copolymer system does not require tailored DNA barcodes or substrates and is compatible with various DNA barcoding strategies. This approach has the potential to significantly expand the range of DEL-compatible synthesis methods and may find applications in other nucleic acid-based technologies requiring a broader solvent scope. Future work will focus on further investigating the hydrophilic–lipophilic balance of the copolymer to improve understanding of the aggregate structure and potentially extend the solvent scope.

Alyssa Grogan , Robin M. Perelli , Seungkirl Ahn , Haoran Jiang , Arun Jyothidasan , Damini Sood , Chongzhao You , David I. Israel , Alex Shaginian , Qiuxia Chen , Jian Liu , Jialu Wang , Jan Steyaert , Alem W. Kahsai , Andrew P. Landstrom , Robert J. Lefkowitz , Howard A. Rockman The Journal of Clinical Investigation  DOI: 10.1172/jci190252 Abstract Orthosteric β-blockers represent the leading pharmacological intervention for managing heart diseases owing to their ability to competitively antagonize β-adrenergic receptors (βARs). However, their use is often limited by the development of adverse effects such as fatigue, hypotension, and reduced exercise capacity, due in part to the nonselective inhibition of multiple βAR subtypes. These challenges are particularly problematic in treating catecholaminergic polymorphic ventricular tachycardia (CPVT), a disease characterized by lethal tachyarrhythmias directly triggered by cardiac β1AR activation. To identify small molecule allosteric modulators of the β1AR that could offer enhanced subtype specificity and robust functional antagonism of β1AR-mediated signaling, we conducted a DNA-encoded small molecule library screen and discovered Compound 11 (C11). C11 selectively potentiates the binding affinity of orthosteric agonists to the β1AR while potently inhibiting downstream signaling following β1AR activation. Moreover, C11 prevents agonist-induced spontaneous contractile activity, Ca2+ release events, and exercise-induced ventricular tachycardia in the CSQ2–/– murine model of CPVT. Collectively, our studies demonstrate that C11 belongs to an emerging class of allosteric modulators termed PAM-antagonists that positively modulate agonist binding but block downstream function. With unique pharmacological properties and selective functional antagonism of β1AR-mediated signaling, C11 represents a promising therapeutic candidate for the treatment of CPVT and other forms of cardiac disease associated with excessive β1AR activation.

Julien Poupart Annual Review of Cancer Biology https://doi.org/10.1146/annurev-cancerbio-071124-011719 Abstract This review describes recent developments in DNA-encoded library (DEL) technology, which has enabled transformative discoveries in cancer research. Successful DEL screening campaigns for cancer-relevant targets are described in detail to highlight the unique advantages of this technology compared to other hit-generation strategies. Moreover, recent developments in screening methods that have helped expand the DEL-addressable target space are described, and their implications for cancer research are emphasized. DEL screening campaigns targeting RNA and transcription factors are discussed, and various cell-based DEL evaluation methods for membrane proteins are compared and put into context. Finally, the use of DEL technology for the discovery of novel bifunctional degraders is presented. Overall, this article provides a comprehensive overview of key DEL discoveries that are expected to be of significant interest to cancer researchers and medicinal chemists working in the field of oncology.

Kangyin Pan , Wentao Meng , Ying Yao , Wanting Bi , Guang Yang , Hongtao Xu Organic Letters  DOI: 10.1021/acs.orglett.5c03695 Abstract The COVID-19 pandemic incited a global health crisis that accelerated the development of antiviral therapeutics. One successful avenue for inhibiting SARS-CoV-2 has been through targeting the main protease (Mpro; 3CLpro), a key enzyme for the viral lifecycle that cleaves at 11 sites in the viral polyprotein pp1a and pp1ab. Although potent inhibitors of Mpro have been discovered, including the FDA-approved drug Paxlovid, the potential emergence of resistant variants requires continued antiviral development efforts. The current methods to characterize binders of Mpro, such as SPR or ITC, are costly and time-consuming. To improve the speed and feasibility of Mpro inhibitor development, we developed a competitive miniaturized fluorescence polarization (FP) binding assay. We repurposed small molecules from a DNA-encoded library screen into FP probes by appending a fluorophore with various linkers. After identifying a probe that exhibited potent Mpro binding (KD ∼43 nM), we optimized buffer conditions, pH, and additives. Assay validation revealed that our competitive fluorescence polarization assay was robust, with a Z′-score of 0.79 and a signal window of 23. This assay can be used as a single-point assay for high-throughput screening or to triage small molecules by generating Ki values for binding. Efforts from this work resulted in an Mpro binding assay that requires minimal protein consumption, low sample volumes, short incubation times (30 min), and operates at room temperature. In conclusion, we developed a robust FP assay that can be used to rapidly characterize Mpro-binding small molecules to support the development of new SARS-CoV-2 antivirals.

Riya, Singh, Aryan Amit, Barsainyan, Abhiraj Pravin, Mengade, Rida, Irfan, Bharath, Ramsundar ChemRxiv DOI: 10.26434/chemrxiv-2025-f11mk Abstract DNA-encoded libraries (DELs) have emerged as a powerful platform for screening ultra-large chemical spaces by leveraging DNA barcodes to tag and track individual small molecules. Recent work has shown that machine learning can enhance DEL based hit discovery by denoising sequencing artifacts and improving binder identification. However, existing tools for DEL modeling remain fragmented, limiting reproducibility and scalability. To address these challenges, we introduce Deepchem-DEL, an open source suite of workflows built on top of the DeepChem ecosystem. Deepchem-DEL integrates (i) a configurable denoising pipeline and (ii) modular Deepchem workflows for enrichment/hit prediction and benchmarking. We evaluated Deepchem-DEL using the KinDEL dataset and reproduced key baselines across diverse model architectures. Our experiments demonstrate that Deepchem-DEL enables reproducible and scalable machine learning workflows for DEL modeling, reducing engineering overhead for hit discovery.

Mackenzie K. Wyllie, Rayhan G. Biswas, Jyoti Vishwakarma, Morgan A. Esler, Joseph A. Rollie, Hideki Aihara, Reuben S. Harris, Daniel A. Harki ACS Pharmacology & Translational Science  DOI: 10.1021/acsptsci.5c00463 Abstract The COVID-19 pandemic incited a global health crisis that accelerated the development of antiviral therapeutics. One successful avenue for inhibiting SARS-CoV-2 has been through targeting the main protease (Mpro; 3CLpro), a key enzyme for the viral lifecycle that cleaves at 11 sites in the viral polyprotein pp1a and pp1ab. Although potent inhibitors of Mpro have been discovered, including the FDA-approved drug Paxlovid, the potential emergence of resistant variants requires continued antiviral development efforts. The current methods to characterize binders of Mpro, such as SPR or ITC, are costly and time-consuming. To improve the speed and feasibility of Mpro inhibitor development, we developed a competitive miniaturized fluorescence polarization (FP) binding assay. We repurposed small molecules from a DNA-encoded library screen into FP probes by appending a fluorophore with various linkers. After identifying a probe that exhibited potent Mpro binding (KD ∼43 nM), we optimized buffer conditions, pH, and additives. Assay validation revealed that our competitive fluorescence polarization assay was robust, with a Z′-score of 0.79 and a signal window of 23. This assay can be used as a single-point assay for high-throughput screening or to triage small molecules by generating Ki values for binding. Efforts from this work resulted in an Mpro binding assay that requires minimal protein consumption, low sample volumes, short incubation times (30 min), and operates at room temperature. In conclusion, we developed a robust FP assay that can be used to rapidly characterize Mpro-binding small molecules to support the development of new SARS-CoV-2 antivirals.

Wei Hou, Shuning Zhang, Wei Yi, Peixiang Ma, Hongtao Xu  Trends in Chemistry https://doi.org/10.1016/j.trechm.2025.08.008 Abstract DNA-encoded libraries (DELs) have emerged as a critical technology for rapid hit identification in drug discovery. The effectiveness of DEL selection fundamentally depends on the chemical diversity present within the library. As DNA-compatible chemistry plays a central role in expanding both the chemical space and molecular diversity of DELs, this area has seen significant advancements in recent years. Additionally, research in DNA-compatible chemistry and bioconjugation chemistry demonstrates a synergistic relationship. This review focuses on the most recent developments in DNA-compatible chemistry and highlights its expanding applications beyond traditional library construction. Current synthetic challenges are examined, and potential directions for future research and development are explored.

James Wellnitz , Shabbir Ahmad , Nabin Bagale , Xuemin Cheng , Jermiah Joseph , Hong Zeng , Albina Bolotokova , Aiping Dong , Shaghayegh Reza , Pegah Ghiabi , Elisa Gibson , Guiping Tu , Xianyang Li , Jian Liu , Dengfeng Dou , Jin Li , Timothy L. Foley , Anthony R. Harris , Jacquelyn L. Klug-McLeod , Jisun Lee , Zsofia Lengyel-Zhand , Justin I. Montgomery , Sylvie Sakata , Jinzhi Zhang , Hongyao Zhu , Dafydd R. Owen , Rachel J. Harding , Aled M. Edwards , Benjamin Haibe-Kains , Levon Halabelian , Alexander Tropsha , Rafael M. Couñago Journal of Medicinal Chemistry   DOI: 10.1021/acs.jmedchem.5c01972   Abstract Machine learning (ML) is increasingly used in DNA-encoded library (DEL) screening for ligand discovery, but its success depends on access to suitable data sets, which are often proprietary and costly. To overcome this, we present the first fully open, automated DEL-ML framework using public DEL data sets and chemical fingerprints to enable reproducible, accessible drug discovery. Our workflow─from model training to virtual screening and compound selection─requires no human intervention. As a proof of concept, we identified binders for WDR91 by training ML models on the HitGen OpenDEL library (3B molecules) and screening the Enamine REAL Space library (37B molecules), yielding 50 candidates. Experimental testing confirmed seven novel binders with dissociation constants between 2.7–21 μM. Our open-source approach matches the performance of proprietary methods, demonstrating that public DEL data can support robust ML-driven ligand discovery and fostering transparency and broader community participation in drug development.

Pratik R. Chheda, Dominic S. Finis, Nicholas Simmons, Zhicai Shi Org. Lett. 2025 https://doi.org/10.1021/acs.orglett.5c03424 Abstract The expansion of chemical diversity in DNA-encoded libraries (DELs) is essential for broadening their potential in drug discovery. Herein, we introduce the extended utility of the Surfactant–DNA (Surf-DNA) workflow that enables on-DNA deoxygenative Csp2–Csp3 cross-coupling of on-DNA halides with activated alcohols via metallaphotoredox catalysis under anhydrous conditions. The developed reaction demonstrates broad substrate scope for both coupling partners, high conversion rates, and proficiency in preserving DNA integrity and allows for the efficient incorporation of alcohol-derived scaffolds, which are an underutilized and widely available class of building blocks.   Summary This study presents a novel method for expanding the chemical diversity in DNA-encoded libraries (DELs) through on-DNA deoxygenative C(sp2)−C(sp3) cross-coupling. The Surfactant−DNA (Surf-DNA) workflow enables this coupling under anhydrous conditions, leveraging metallaphotoredox catalysis. The method shows high conversion rates and broad substrate scope, including primary, secondary, and multifunctional alcohols, as well as various (hetero)aryl halides. The study demonstrates the preservation of DNA integrity and the potential for incorporating a wide range of alcohol-derived scaffolds, significantly broadening the chemical space accessible in DEL synthesis.   Highlights - Introduction of a Surfactant−DNA (Surf-DNA) workflow for on-DNA deoxygenative C(sp2)−C(sp3) cross-coupling. - Utilization of metallaphotoredox catalysis under anhydrous conditions. - Demonstration of broad substrate scope and high conversion rates for both coupling partners. - Preservation of DNA integrity and compatibility with elongated DNA substrates. - Efficient incorporation of alcohol-derived scaffolds, expanding the chemical diversity in DELs.   Conclusion The study successfully developed a robust protocol for on-DNA deoxygenative coupling via the Surf-DNA workflow. This method enables high efficiency, expansive substrate scope, functional group tolerance, and preservation of DNA integrity. By unlocking the potential of a widely available and historically underutilized class of alcohol building blocks, this approach allows for the construction of challenging C(sp2)−C(sp3) bonds and the generation of more structurally and functionally diverse libraries. This significantly broadens the chemical space accessible in DEL synthesis, enhancing their potential in drug discovery.

Tao Chen, Longying Cai, Xiaofei Dong, Lifang Zhang, Xuemin Cheng, Jingsong Qu, Guanyu Yang, Sen Gao, Linfu Luo, Huiyong Ma, Shuai Xia, Guansai Liu, Jin Li, Jianyou Shi, Dengfeng Dou  Bioorganic & Medicinal Chemistry https://doi.org/10.1016/j.bmc.2025.118448 Abstract To better understand how pre-installed covalent warheads affect the ligand discovery in DNA-encoded libraries (DELs), three individual covalent DELs incorporating 7, 32, and 64 cysteine-targeting covalent warheads, respectively, were designed and screened against JAK3 purified protein. The experiments resulted in 6 novel series of covalent inhibitors with drug-like properties, with the most potent compounds achieving picomolar IC50 and good selectivity against a mini panel of kinases. Mass spectrometry studies confirmed their covalent mechanisms of action (MOAs) by targeting JAK3 Cys909. Importantly, the synergistic effect of the binding moiety and warhead was confirmed by comparing the activities with their close analogs, suggesting that these compounds may not be designed by simply installing covalent warheads onto reversible binders. Further analysis revealed that 7 warheads were sufficient for identifying JAK3 covalent ligands. This work deepens our understanding of the design and screening of covalent DELs and demonstrates the power of DEL in the identification of diverse covalent inhibitors.   Summary This study explores the impact of covalent warheads in DNA-encoded libraries (DELs) on ligand discovery by designing and screening three covalent DELs with varying numbers of cysteine-targeting warheads against JAK3. The results identified six novel series of covalent inhibitors with drug-like properties, achieving picomolar IC50 values and good selectivity. Mass spectrometry confirmed the covalent binding to JAK3 Cys909. The study highlights the necessity of covalent warheads for effective screening and demonstrates that a library with 7 warheads was sufficient for hit identification. This work provides insights into optimizing covalent DEL design and screening strategies for discovering potent and selective covalent inhibitors.   Highlights - Covalent DNA-encoded libraries (CoDELs) with 7, 32, and 64 cysteine-targeting warheads were designed and screened against JAK3. - Six novel series of covalent inhibitors with picomolar IC50 values and good selectivity were discovered. - Mass spectrometry confirmed covalent binding to JAK3 Cys909. - The study demonstrated the necessity of covalent warheads for effective screening and identified 7 warheads as sufficient for hit identification. - The findings provide a robust framework for optimizing covalent DEL design and screening strategies.   Conclusion This research demonstrates the power of DNA-encoded library (DEL) technology in discovering novel covalent inhibitors for JAK3. By incorporating covalent warheads into DELs, we identified six series of potent and selective covalent inhibitors with drug-like properties. The study highlights the importance of covalent warheads in achieving effective screening and confirms that a library with 7 distinct warheads is sufficient for identifying JAK3 covalent ligands. The findings provide valuable insights into optimizing covalent DEL design and screening strategies, paving the way for the discovery of novel covalent inhibitors for other challenging targets.  

Rui Jin and Xiaojie Lu https://doi.org/10.1016/j.sbi.2025.103163 Abstract DNA-encoded library (DEL) technology has enabled efficient discovery of both non-covalent and covalent inhibitors. Covalent DELs (CoDELs) incorporating diverse electrophilic warheads have expanded the scope of covalent targeting beyond cysteine to residues like lysine, tyrosine, arginine, and glutamic acid. The integration of CoDEL with activity-based protein profiling (ABPP) has further enabled the identification of potential protein targets for CoDEL screening using residue-selective warheads. Additionally, proteome profiling with fully functionalized tags has guided target identification for focused DELs with privileged structures. This review highlights recent advances in CoDEL technologies for targeting both cysteine and non-cysteine residues and discusses how proteomics facilitates hit discovery through CoDELs and focused DELs.   Summary This review article discusses the recent advancements in DNA-encoded libraries (DELs), particularly focusing on covalent DELs (CoDELs). CoDELs have shown significant potential in discovering covalent inhibitors targeting various amino acid residues, not just cysteine. The article highlights the use of CoDELs in identifying inhibitors for a range of targets, including viral proteins and oncoproteins. It also explores the integration of CoDEL technology with proteomics strategies such as activity-based protein profiling (ABPP) and proteome profiling using fully functionalized tags to guide the construction of focused DELs and improve hit discovery. The review emphasizes the importance of these integrative approaches in expanding the scope of covalent drug design and accelerating the discovery of novel therapeutic agents.   Highlights - Covalent DNA-encoded libraries (CoDELs) have expanded covalent targeting to residues beyond cysteine, including lysine, tyrosine, arginine, and glutamic acid. - Integration of CoDEL with activity-based protein profiling (ABPP) enables the identification of potential protein targets for covalent inhibitors. - Proteome profiling using fully functionalized tags guides the construction of focused DELs with privileged structures, enhancing hit discovery. - Recent studies demonstrate the discovery of covalent inhibitors for viral proteins and oncoproteins using CoDELs, showcasing their translational potential. - The review emphasizes the importance of combining proteomics with CoDEL technology to improve target selection and hit identification.   Conclusion Over the past decade, DNA-encoded library (DEL) technology has undergone rapid development, with significant advancements in the expansion of on-DNA chemical reactions and the construction of diverse libraries. Covalent DELs (CoDELs) have emerged as a powerful tool for discovering covalent inhibitors targeting various amino acid residues. The integration of CoDEL with proteomics strategies such as activity-based protein profiling (ABPP) and proteome profiling using fully functionalized tags has significantly enhanced the efficiency of hit discovery and target identification. As the field progresses, the continued expansion of CoDEL applications and the incorporation of proteomic data to guide library design and target engagement represent promising frontiers for both covalent and non-covalent DELs, potentially leading to the discovery of novel therapeutic agents with improved efficacy and selectivity.

Artur Menzeleev, Sathya Chitturi, Geraint Davies ,Tony Schroeder ,Alpha Lee ChemRxiv  D O I: 10.26434/chemrxiv-2025-8rw8j Abstract DNA-encoded libraries (DELs) are a powerful way to find chemical starting points against challenging biological targets, by rapidly generating billion-scale structure-activity datasets. However, DEL experiment design and interpretation, especially the optimal use of machine learning (ML) to analyse the vast amount of generated data or to screen large external purchasable datasets, remain poorly understood. To address these challenges, we report the development of a digital twin – an in-silico DEL simulator – that models the underlying chemistry and selection processes of typical experiments as a function of key design parameters, including read count, cycles of selection, one-step reaction yield, and library size. We systematically investigate how these design parameters influence downstream ML virtual screening and identify specific regimes where the choice to apply preprocessing steps such as disynthon aggregation can significantly enhance screening performance. In addition, we show that increasing library size can degrade ML-based screening performance. Our simulator provides a statistically principled way to understand and analyse DEL experiments via an interpretable model for DEL data generation.

Marcos E. Milla; Jonathan M. Blevitt*; Steven D. Goldberg;Anthony A. Armstrong;Katherine Y. Blain;Krystal L. Herman;Annie X. Liu;Rosa Luna;Cynthia M. Milligan;Aaron N. Patrick;Ruth A. Steele;Scott D. Bembenek;Paolo Centrella;Matthew A. Clark;John W. Cuozzo;Jeremy S. Disch;Derrick Domingo;Avery Hunt;Christoper D. Hupp;Anthony D. Keefe;Jinquan Luo;Tara Mirzadegan;Marina I. Nelen;Daniel I. Resnicow;Eric A. Sigel;Holly H. Soutter;Dawn M. Troast;Xiaohua XueFang Yi;Ying Zhang;Paul F. Jackson;James P. Edwards;Kevin J. Lumb ACS Med. Chem. Lett. 2025, XXXX, XXX, XXX-XXX https://doi.org/10.1021/acsmedchemlett.5c00502 Abstract A novel series of inhibitors of the interaction of IL-17A with its cognate receptor has been discovered using DNA-encoded library (DEL) technology. The lead compound (JNJ627, Compound 1) of the series occupies the interior interface of the IL-17A homodimer and disables receptor binding. The mechanism of action involves allosteric disruption of the IL-17A quaternary structure to prevent adoption of the receptor-binding conformation, rather than direct orthosteric inhibition at the receptor-binding site. Molecules of this series exhibit remarkably slow on-rate kinetics and potent inhibition of IL-17A signaling in human primary cells.

Suraj Kanoo,∥ Eduardo de Pedro Beato,∥ Tim Schulte, Lara Vogelsang, Luca Torkowski, Felix Waldbach, Philipp Hartmann, Riya Kayal, Karl-Josef Dietz, and Tobias Ritter* J. Am. Chem. Soc. 2025, https://doi.org/10.1021/jacs.5c11842   Abstract C–N cross coupling reactions are widely employed for the construction of carbon–nitrogen bonds. However, control of chemoselectivity in the presence of the amino functionality in oligonucleotides remains a challenge. Here, we report the development of a new ruthenium reagent that enables the chemoselective N-arylation of amine–DNA conjugates with distinct chemoselectivity when compared to conventional palladium-based C–N bond-forming catalysts. The ruthenium reagent activates commercially available haloarenes in situ via η6 π-arene coordination for subsequent SNAr with the amine. The method is compatible with various commercially available haloarenes and aliphatic amines, and the reaction proceeds under mild conditions.

Gavin W. Collie*, Bryony Ackroyd, Catriona Corbishley, Daniel H. O’Donovan, Alex Edwards, Andrea Gohlke, Xiaoxiao Guo, Bethan Howells, Yuliang Li, Andrew Madin, Alexander G. Milbradt, Emma L. Rivers*,Sandeep K. Talapatra, Elizabeth Underwood, and Alice Webb ACS Med. Chem. Lett.  DOI: 10.1021/acsmedchemlett.5c00396   Abstract NSD2 is a key epigenetic regulator and has received considerable attention as a drug target due to its well-documented role in tumorigenesis. We report here a DNA-encoded library screen targeting the PWWP1 domain of NSD2 from which we discovered novel, potent, and selective binders. Furthermore, these compounds were used to develop a novel crystal system, increasing our understanding of the folding of this domain. Together, these results provide a solid molecular and structural basis for the further study of the PWWP1 domain of NSD2 as a cancer drug target.

Tony Georgiev, Francesca Migliorini, Andrea Ciamarone, Marco Mueller, Ilaria Biancofiore, Pinuccia Faviana, Francesco Bartoli, Young Seo Park Kim, Lucrezia Principi, Ettore Gilardoni, Gabriele Bassi, Nicholas Favalli, Emanuele Puca, Dario Neri, Sebastian Oehler & Samuele Cazzamalli  Nature Biomedical Engineering (2025) DOI: 10.1038/s41551-025-01432-6   Abstract Improving the specificity of prostate cancer treatment requires ligands that bind selectively and with ultra-high affinity to tumour-associated targets absent from healthy tissues. Prostatic acid phosphatase has emerged as an alternative target to prostate-specific membrane antigen, as it is expressed in a broader subset of prostate cancers and is not detected in healthy organs such as the salivary glands and kidneys. Here, to discover selective binders to prostatic acid phosphatase, we constructed two DNA-encoded chemical libraries comprising over 6.7 million small molecules based on proline and phenylalanine scaffolds. Screening against the purified human prostatic acid phosphatase yielded OncoACP3, a small organic ligand with picomolar binding affinity. When radiolabelled with lutetium-177, OncoACP3 selectively accumulated in enzyme-expressing tumours with a long residence time (biological half-life greater than 72 h) and a high tumour-to-blood ratio (>148 at 2 h after administration). Lutetium-177-labelled OncoACP3 cured tumours in mice at low, well-tolerated doses. Its conjugation to the cytotoxic agent monomethyl auristatin E facilitated tumour-selective payload deposition, resulting in potent anti-tumour activity. The modular structure of OncoACP3 supports flexible payload delivery for the targeted treatment of metastatic prostate cancer.   Summary Philochem’s research team successfully identified high-affinity ligands targeting prostate-specific membrane antigen (ACP3) using DNA-encoded library (DEL) technology. The study demonstrates a rapid and efficient path from hit identification to preclinical validation, highlighting DEL’s utility in accelerating radioligand therapy development.   Highlights 1. High-Affinity Ligand Discovery - Two phosphonate-focused DELs were screened against ACP3, yielding enriched hits with strong binding motifs. - Optimized compounds achieved sub-nanomolar inhibition (SPR-confirmed) and >100-fold improved affinity versus the original ligand. - Fluorophore-conjugated ligands selectively stained ACP3-expressing prostate cancer cells, confirming target engagement. 2. Therapeutic Efficacy in Preclinical Models - 177Lu-labeled conjugates showed ~70 %ID/g tumor uptake in xenografts with minimal off-target accumulation and slow washout. - Significant tumor regression was observed, outperforming a reference radioligand derived from earlier inhibitors. - Small-molecule drug conjugates (cleavable linker + MMAE payload) also demonstrated potent antitumor activity.   Conclusion Philochem’s work delivers a robust pipeline of ACP3-targeting ligands with translational potential in radioligand therapy and antibody-free drug conjugates. It validates DEL as a key enabling technology for accelerating cancer therapeutic discovery.