Large Language Model is Secretly a Protein Sequence Optimizer

Yinkai Wang; Jiaxing He; Yuanqi Du; Xiaohui Chen; Jianan Canal Li; Li-Ping Liu; Xiaolin Xu; Soha Hassoun

Large Language Model is Secretly a Protein Sequence Optimizer

Machine Learning 2025-01-20 v2 Artificial Intelligence Quantitative Methods

Authors: Yinkai Wang , Jiaxing He , Yuanqi Du , Xiaohui Chen , Jianan Canal Li , Li-Ping Liu , Xiaolin Xu , Soha Hassoun

View on arXiv ↗ PDF ↗

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.

Keywords

evolutionary optimization large language model instruction tuning

Cite

@article{arxiv.2501.09274,
  title  = {Large Language Model is Secretly a Protein Sequence Optimizer},
  author = {Yinkai Wang and Jiaxing He and Yuanqi Du and Xiaohui Chen and Jianan Canal Li and Li-Ping Liu and Xiaolin Xu and Soha Hassoun},
  journal= {arXiv preprint arXiv:2501.09274},
  year   = {2025}
}

Comments

Preprint

Large Language Model is Secretly a Protein Sequence Optimizer

Abstract

Keywords

Cite

Comments

Related papers