AgentGC: Evolutionary Learning-based Lossless Compression for Genomics Data with LLM-driven Multiple Agent

2026-01-20

AI Summary

This work addresses the limitations of existing lossless genomic compression methods—such as shallow modeling, poor adaptability, and unfriendly user interaction—by proposing a novel three-tier multi-agent evolutionary compression framework powered by large language models (LLMs). The user layer enables natural language interaction, the cognitive layer jointly optimizes algorithmic, data, and system configurations, and the compression layer automates multi-knowledge learning to perform compression and decompression, offering three operational modes: compression-ratio priority, throughput priority, and balanced. To our knowledge, this is the first integration of LLMs and multi-agent evolutionary learning into genomic compression, yielding an evolvable, adaptive, and user-friendly system. Evaluated across nine datasets against 14 baseline methods, the framework achieves an average compression ratio improvement of 16.37% and up to a 9.23× increase in throughput.

Technology Category

• Search and Optimization: Evolutionary Computation
• Machine Learning: Evolutionary Learning
• Data Mining & Knowledge Management: Data Compression

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Lossless compression has made significant advancements in Genomics Data (GD) storage, sharing and management. Current learning-based methods are non-evolvable with problems of low-level compression modeling, limited adaptability, and user-unfriendly interface. To this end, we propose AgentGC, the first evolutionary Agent-based GD Compressor, consisting of 3 layers with multi-agent named Leader and Worker. Specifically, the 1) User layer provides a user-friendly interface via Leader combined with LLM; 2) Cognitive layer, driven by the Leader, integrates LLM to consider joint optimization of algorithm-dataset-system, addressing the issues of low-level modeling and limited adaptability; and 3) Compression layer, headed by Worker, performs compression&decompression via a automated multi-knowledge learning-based compression framework. On top of AgentGC, we design 3 modes to support diverse scenarios: CP for compression-ratio priority, TP for throughput priority, and BM for balanced mode. Compared with 14 baselines on 9 datasets, the average compression ratios gains are 16.66%, 16.11%, and 16.33%, the throughput gains are 4.73x, 9.23x, and 9.15x, respectively.

Problem

• lossless compression
• genomics data
• evolutionary learning
• limited adaptability
• user-unfriendly interface

Innovation

• evolutionary learning
• multi-agent system
• LLM-driven compression
• genomics data
• lossless compression