Assay2Mol: large language model-based drug design using BioAssay context

2025-07-16

AI Summary

Early-stage drug discovery faces a bottleneck in de novo molecular generation for novel targets with no experimentally resolved protein structures. Method: We propose a new generative paradigm that jointly leverages unstructured bioassay text (e.g., mechanistic descriptions, experimental protocols) and quantitative high-throughput screening (HTS) data—without requiring 3D target structures. Using large language models (LLMs), we parse assay narratives and integrate them with HTS outcomes from structurally similar targets; context-aware prompting then drives molecule generation under synthetic accessibility constraints. Contribution/Results: To our knowledge, this is the first approach to jointly model descriptive bioassay text and quantitative HTS data for structure-agnostic molecular design. In benchmarking, our method significantly outperforms existing ML-based generative models, yielding molecules with improved chemical validity (SA Score), synthetic tractability (SAscore), and target relevance—establishing a scalable, data-driven framework for novel-target-oriented de novo drug design.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Scientific databases aggregate vast amounts of quantitative data alongside descriptive text. In biochemistry, molecule screening assays evaluate the functional responses of candidate molecules against disease targets. Unstructured text that describes the biological mechanisms through which these targets operate, experimental screening protocols, and other attributes of assays offer rich information for new drug discovery campaigns but has been untapped because of that unstructured format. We present Assay2Mol, a large language model-based workflow that can capitalize on the vast existing biochemical screening assays for early-stage drug discovery. Assay2Mol retrieves existing assay records involving targets similar to the new target and generates candidate molecules using in-context learning with the retrieved assay screening data. Assay2Mol outperforms recent machine learning approaches that generate candidate ligand molecules for target protein structures, while also promoting more synthesizable molecule generation.

Problem

• Leveraging unstructured biochemical assay text for drug discovery
• Generating candidate molecules using assay context and LLMs
• Improving synthesizability and performance over structure-based ML methods

Innovation

• Uses large language model for drug design
• Retrieves similar assay records for targets
• Generates synthesizable candidate molecules