Fine-tuning Boltz-2 with few data improves early drug candidate ranking
Adapting Boltz-2 with limited experimental activity data improves early enrichment in virtual screening
Artificial IntelligenceMachine Learning
Summary
Finding promising drug compounds from huge chemical libraries is expensive and slow. The authors show that carefully adjusting a computer model called Boltz-2 using a small amount of experimental data can better identify likely active compounds early on. This adjustment method works well even when only a limited number of test results are available. They also demonstrate that rechecking only the top candidates keeps the benefit while reducing extra work.
What this means in practice
- •For drug discovery teams: Improve prioritization of active compounds in drug screening with minimal new experiments by fine-tuning Boltz-2.
- •For biotech screening labs: Cut workload by rescoring only top candidates while retaining improved hit identification from virtual screens.
Authors
Kairi Furui, Masahito Ohue
Abstract
Virtual screening aims to prioritize active compounds from large chemical libraries within a limited experimental budget. When applying Boltz-2 to virtual screening, a key challenge is how to use limited experimental data from the target assay to improve the prioritization of active compounds. We investigated whether fine-tuning the Boltz-2 affinity heads with a small number of binary activity labels could improve early enrichment of active compounds in hit discovery. We compared fine-tuning with 40-300 labels in a retrospective evaluation on eight MF-PCBA targets. With 300 activity measurements, fine-tuning increased the number of actives in the top 1% by a geometric mean of 1.77-fold across the eight targets and improved average precision (AP) by 2.14-fold relative to the control without fine-tuning. We also investigated whether rescoring a subset of candidates could retain the improvement in hit recovery by reranking only the top-ranked Boltz-2 candidates with the fine-tuned head. Restricting rescoring to approximately 10% of the evaluation set retained hit recovery comparable to full rescoring. These findings show that affinity-head fine-tuning with limited activity labels improves early enrichment with Boltz-2 and that this benefit can be retained when rescoring a restricted set of candidates.