Large Language Model is Secretly a Protein Sequence Optimizer

2025/01/16 by Yinkai Wang, Wang, Yinkai, Jiaxing He +15 · 2 voices
Biochemistry, Genetics and Molecular Biology · Computer Science · #Machine Learning in Bioinformatics #Topic Modeling #cs.AI #cs.LG #q-bio.QM

paper · pdf · doi:10.48550/arxiv.2501.09274

openalex publication_date 2025/01/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We consider the protein sequence engineering problem, which aims to find protein sequences with high fitness levels, starting from a given wild-type sequence. Directed evolution has been a dominating paradigm in this field which has an iterative process to generate variants and select via experimental feedback. We demonstrate large language models (LLMs), despite being trained on massive texts, are secretly protein sequence optimizers. With a directed evolutionary method, LLM can perform protein engineering through Pareto and experiment-budget constrained optimization, demonstrating success on both synthetic and experimental fitness landscapes.

Discussions

• 🧪 Researchers show that standard LLMs, like Llama-3-1-Instruct, can optimize protein engineering tasks without modification, outperforming evolutionary baselines. 🧬💻 #MLSky [bsky, 13 points, 0 comments]
• In arxiv.org/pdf/2501.09274 a standard LLM w/o fine tuning is used to successfully predict protein sequences fitter for certain purposes (not specified specifically!) by guiding mutations in an evolut [bsky, 0 points, 2 comments]

Related

• Boosting In-Silicon Directed Evolution with Fine-Tuned Protein Language Model and Tree Search
• Protein Language Model Fitness Is a Matter of Preference
• Training Compute-Optimal Protein Language Models
• Training Compute-Optimal Protein Language Models
• Structure-informed Language Models Are Protein Designers