From Single to Multi-Agent Reasoning: Advancing GeneGPT for Genomics QA

2026-01-15

AI Summary

Efficiently and accurately extracting information from complex, fragmented genomic databases to support biomedical question answering remains challenging, as existing approaches are constrained by rigid API dependencies and limited adaptability. This work proposes GenomAgent, a multi-agent framework tailored for genomic question answering, which coordinates multiple specialized agents through task decomposition and collaborative mechanisms to flexibly invoke domain-specific APIs and effectively leverage structured knowledge bases. By overcoming the limitations of single-model architectures and fixed interfaces, GenomAgent achieves an average performance improvement of 12% over GeneGPT across nine tasks on the GeneTuring benchmark, significantly enhancing answer accuracy and system scalability while demonstrating promising potential for transfer to other scientific domains.

Technology Category

• Multiagent Systems: Modeling other Agents
• Natural Language Processing: Question Answering
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Comprehending genomic information is essential for biomedical research, yet extracting data from complex distributed databases remains challenging. Large language models (LLMs) offer potential for genomic Question Answering (QA) but face limitations due to restricted access to domain-specific databases. GeneGPT is the current state-of-the-art system that enhances LLMs by utilizing specialized API calls, though it is constrained by rigid API dependencies and limited adaptability. We replicate GeneGPT and propose GenomAgent, a multi-agent framework that efficiently coordinates specialized agents for complex genomics queries. Evaluated on nine tasks from the GeneTuring benchmark, GenomAgent outperforms GeneGPT by 12% on average, and its flexible architecture extends beyond genomics to various scientific domains needing expert knowledge extraction.

Problem

• genomic question answering
• large language models
• domain-specific databases
• API dependencies
• multi-agent reasoning

Innovation

• multi-agent framework
• genomic question answering
• large language models
• expert knowledge extraction
• flexible architecture