Conditional Chemical Language Models are Versatile Tools in Drug Discovery

TL;DR

SAFE-T introduces a versatile conditional chemical language model that effectively integrates biological context for molecular design, scoring, and interpretability, significantly advancing drug discovery processes.

Contribution

The paper presents SAFE-T, a novel framework that conditions on biological context for molecule design and scoring without structural data or engineered scores.

Findings

Achieves competitive or superior performance on multiple benchmarks.

Supports goal-directed molecule generation aligned with biological objectives.

Provides interpretable insights into structure-activity relationships.

Abstract

Generative chemical language models (CLMs) have demonstrated strong capabilities in molecular design, yet their impact in drug discovery remains limited by the absence of reliable reward signals and the lack of interpretability in their outputs. We present SAFE-T, a generalist chemical modeling framework that conditions on biological context -- such as protein targets or mechanisms of action -- to prioritize and design molecules without relying on structural information or engineered scoring functions. SAFE-T models the conditional likelihood of fragment-based molecular sequences given a biological prompt, enabling principled scoring of molecules across tasks such as virtual screening, drug-target interaction prediction, and activity cliff detection. Moreover, it supports goal-directed generation by sampling from this learned distribution, aligning molecular design with biological objectives. In comprehensive zero-shot evaluations across predictive (LIT-PCBA, DAVIS, KIBA, ACNet) and generative (DRUG, PMO) benchmarks, SAFE-T consistently achieves performance comparable to or better than existing approaches while being significantly faster. Fragment-level attribution further reveals that SAFE-T captures known structure-activity relationships, supporting interpretable and biologically grounded design. Together with its computational efficiency, these results demonstrate that conditional generative CLMs can unify scoring and generation to accelerate early-stage drug discovery.