AutoProteinEngine: A Large Language Model Driven Agent Framework for Multimodal AutoML in Protein Engineering

TL;DR

AutoProteinEngine (AutoPE) is a novel agent framework that uses large language models to enable biologists to perform multimodal AutoML for protein engineering, simplifying complex DL tasks through natural language interaction.

Contribution

AutoPE integrates LLMs with AutoML for protein sequence and graph modeling, automatic hyperparameter tuning, and data retrieval, making DL accessible to non-experts in protein engineering.

Findings

AutoPE outperforms traditional zero-shot methods in protein engineering tasks.

AutoPE enables non-experts to effectively utilize DL models.

AutoPE demonstrates significant performance improvements in real-world applications.

Abstract

Protein engineering is important for biomedical applications, but conventional approaches are often inefficient and resource-intensive. While deep learning (DL) models have shown promise, their training or implementation into protein engineering remains challenging for biologists without specialized computational expertise. To address this gap, we propose AutoProteinEngine (AutoPE), an agent framework that leverages large language models (LLMs) for multimodal automated machine learning (AutoML) for protein engineering. AutoPE innovatively allows biologists without DL backgrounds to interact with DL models using natural language, lowering the entry barrier for protein engineering tasks. Our AutoPE uniquely integrates LLMs with AutoML to handle model selection for both protein sequence and graph modalities, automatic hyperparameter optimization, and automated data retrieval from protein databases. We evaluated AutoPE through two real-world protein engineering tasks, demonstrating substantial performance improvements compared to traditional zero-shot and manual fine-tuning approaches. By bridging the gap between DL and biologists' domain expertise, AutoPE empowers researchers to leverage DL without extensive programming knowledge. Our code is available at https://github.com/tsynbio/AutoPE.