ProteinGPT: Multimodal LLM for Protein Property Prediction and Structure Understanding

Yijia Xiao, Edward Sun, Yiqiao Jin, Qifan Wang, Wei Wang

Track: Main track (up to 8 pages)

Abstract: Understanding biological processes, drug development, and biotechnological advancements requires a detailed analysis of protein structures and functions, a task that is inherently complex and time-consuming in traditional protein research. To streamline this process, we introduce ProteinGPT, a state-of-the-art multimodal large language model for proteins, which allows users to upload protein sequences and/or structures for comprehensive proteins analysis and responsive inquiries. ProteinGPT seamlessly integrates protein sequence and structure encoders with linear projection layers to ensure precise representation adaptation. It leverages a large language model (LLM) to generate accurate and contextually relevant responses. To train ProteinGPT, we construct a large-scale dataset of 132,092 proteins, each annotated with 20-30 property tags and 5-10 QA pairs per protein, and optimized the instruction-tuning process using GPT-4o. Experiments demonstrate that ProteinGPT effectively generates informative responses to protein-related questions, achieving high performance on both semantic and lexical metrics. It significantly outperforms baseline models and general-purpose LLMs in understanding and responding to protein-related queries.

Submission Number: 7