Exploring the Scoring Function Space with Lasso Regression.

Fiche publication

Date publication

janvier 2026

Journal

Methods in molecular biology (Clifton, N.J.)

Auteurs

Membres identifiés du Cancéropôle Est :
Pr BAUD Stéphanie

Tous les auteurs :
da Silva AD, Baud S, de Azevedo WF

Lien Pubmed

https://www.ncbi.nlm.nih.gov/pubmed/41075082

Résumé

Artificial intelligence (AI) successfully integrates several emerging and established techniques to build models to address complex systems, including those from biological sources. In developing novel technologies to address protein-ligand interactions, AI showed relevant results for the structural modeling of protein targets (e.g., AlphaFold) and for building new scoring functions to address intermolecular interactions. Analysis of protein-ligand interactions is central to any docking screen project, and these AI developments have great potential to contribute to speeding up drug discovery and increasing the reliability of the computational methods employed to study intermolecular interactions. In this chapter, we present the Lasso regression method available in the program SAnDReS 2.0 and discuss its application to build a regression model to predict the inhibition of a protein target used in developing anticancer drugs. We explain the scoring function concept to get further insights into developing models to predict binding affinity. We focused our discussions on open-source software and freely accessible databases to build our regression models. Also, we made available all the codes discussed here at GitHub: https://github.com/azevedolab/docking#readme .