Simon H. Rüdisser, Gabriela Stadler, Alvar D. Gossert

Angewandte Chemie International Edition

DOI: 10.1002/anie.6036832

Abstract

Small‐molecule drug discovery typically begins with the screening of compound libraries to identify initial hits, which are subsequently optimized into lead compounds and, ultimately, drug candidates. Diverse screening methodologies are employed, including DNA‐encoded library technology, high‐throughput screening, and fragment‐based drug discovery (FBDD). Among these, FBDD is particularly powerful when integrated with structure‐guided drug design and biophysical affinity measurements. However, accurately quantifying the weak binding affinities of fragments remains a significant challenge. To address this, we introduce chemical shift anisotropy (CSAKD), a novel method for determining absolute fragment affinities using NMR relaxation. The CSAKD approach eliminates the need for titration experiments and isotopic labeling. Furthermore, we complement this method with a machine learning model for the rapid and accurate prediction of chemical shielding tensors. In summary, CSAKD allows fast and efficient affinity determination which seamlessly integrates into FBDD by NMR.