Support Vector Machine – a model for QSAR still robust in the era of Deep Learning
The new QSAR framework in Flare makes generating and using 3D-QSAR descriptors to build models with good predictivity and generalizability ...
News
A group of academics from Yonsei University, CHA University and and Ewha Womens’ University in South Korea have recently published an interesting paper in Chemical Communications looking at p300 histone acetyltransferase inhibitors as anti-fibrotics. The authors took an initial in-house set of weak binders to the histone acetyltransferase (HAT) region of the target, aligned them using FieldTemplater™ and performed an initial SAR analysis using Activity Miner™ and Activity Atlas™ (all components of Forge™, our ligand-based workbench for molecule design and SAR analysis). A set of compounds on a chalcone scaffold were designed against the SAR findings from this analysis. The most active of these matched the Activity Atlas recommendations well, and was also found to be superior to known p300 inhibitors in both in vitro and in vivo assays, as well as possessing good pharmacokinetic properties. This paper demonstrates the power of advanced ligand-based SAR analysis, especially on less well-understood targets.