InstructPro2 designs proteins from natural language and biology context
InstructPro2: Natural Language and Biological Grounding Guided De Novo Protein Design
Machine Learning
Summary
Designing proteins that do specific jobs is very important but hard. The authors created InstructPro2, a system that turns regular text instructions about what a protein should do into actual protein sequences. It uses information about important biological parts like ligands or antigens to make better designs. Their system works better than previous methods in making proteins that bind targets or act as antibodies.
What this means in practice
- •For biotech engineers: Generate custom proteins that bind specific molecules by interpreting text instructions and biological context simultaneously.
- •For drug development teams: Design antibody proteins optimized for target docking using natural language-guided protein creation.
Authors
Charles Novak, Zhenqiao Song, Chuan Li, Jianwen Xie, Lei Li
Abstract
The ability to generate proteins with desired functions is a central goal of computational biology. Although natural language can express diverse design objectives, existing protein generation methods often support only narrow tasks or make limited use of biological constraints. We develop InstructPro2 to translate natural-language design objectives into protein sequences while incorporating task-specific biological context. A text encoder captures the design instructions, while dedicated ligand and antigen encoders provide biological context to the sequence-generating decoder. For pretraining, we assemble InstructPro2Bench, comprising 205.7 million <function description, protein> pairs from UniProtKB. We subsequently adapt the model to ligand-binding protein and antibody design. On functional protein generation benchmarks, InstructPro2 outperforms Pinal, the strongest baseline, achieving a functional recovery F1 of 74.8 versus 62.2 and lowering global RMSD from 5.33A to 3.04A. For antibody design, the model obtains a DockQ score of 0.39, while ligand-binding protein design reaches a docking success rate of 72.6%. These findings highlight how integrating language-based instructions with biological conditioning can extend protein generation to distinct design settings.