Can AI-Predicted Complexes Teach Machine Learning to Compute Drug Binding Affinity?

Wei-Tse Hsu

doi:10.1021/acs.jcim.5c01848

Can AI-Predicted Complexes Teach Machine Learning to Compute Drug Binding Affinity?

2025-12-01

0

PRE

AI

W

Wei-Tse Hsu

S

Savva Grevtsev

A

Anna M. Herz

T

Thomas Douglas

A

Aniket Magarkar *

P

Philip C. Biggin *

DOI：10.1021/acs.jcim.5c01848

Original

Original request for help

Save

Abstract

En 中文

We evaluate the feasibility of using co-folding models for synthetic data augmentation in training machine learning-based scoring functions (MLSFs) for binding affinity prediction. Our results show that performance gains depend critically on the structural quality of augmented data. In light of this, we established simple heuristics for identifying high-quality co-folding predictions without reference structures, enabling them to substitute for experimental structures in MLSF training. Our study informs future data augmentation strategies based on co-folding models.

Keywords:

LIGAND

Can AI-Predicted Complexes Teach Machine Learning to Compute Drug Binding Affinity?

Abstract

Journal

Organization