Computational Method for Structure-Based Analysis of SAR Transfer

• Published in: Journal of medicinal chemistry Vol. 63; no. 3; pp. 1388 - 1396
• Main Authors: Bonanni, Davide, Lolli, Marco L, Bajorath, Jürgen
• Format: Journal Article
• Language: English
• Published: United States American Chemical Society 13.02.2020
• ISSN: 0022-2623; 1520-4804
• DOI: 10.1021/acs.jmedchem.9b01931

The identification of different compound series with corresponding structure–activity relationship (SAR) progression for a given target is referred to as SAR transfer, which is of interest in lead optimization. If difficulties are encountered during multiproperty optimization, the SAR transfer concept can be applied attempting to replace a lead compound with another candidate. For a systematic assessment of SAR transfer, computational approaches are required. So far, SAR transfer has been investigated at the level of compounds and analogue series. Herein, we introduce a new computational method for structure-guided exploration of SAR transfer. The approach relies on a three-dimensional molecular fragmentation and recombination scheme and the identification of analogues of crystallographic ligands. On the basis of spatially aligned X-ray ligands, alternative substituents and compound cores are identified, enabling the detection of multiple SAR transfer events. Application of the methodology across different targets identified SAR transfer events with high frequency.