PhageBench: Can LLMs Understand Raw Bacteriophage Genomes?

TL;DR

PhageBench is a new benchmark to evaluate large language models' ability to interpret raw bacteriophage genomes, revealing their strengths and limitations in biological reasoning tasks.

Contribution

This work introduces PhageBench, the first benchmark for assessing LLMs on phage genome understanding across multiple bioinformatics tasks.

Findings

LLMs outperform random baselines in phage contig identification.

LLMs show promise in host prediction tasks.

Significant limitations remain in complex reasoning and functional localization.

Abstract

Bacteriophages, often referred to as the dark matter of the biosphere, play a critical role in regulating microbial ecosystems and in antibiotic alternatives. Thus, accurate interpretation of their genomes holds significant scientific and practical value. While general-purpose Large Language Models (LLMs) excel at understanding biological texts, their ability to directly interpret raw nucleotide sequences and perform biological reasoning remains underexplored. To address this, we introduce PhageBench, the first benchmark designed to evaluate phage genome understanding by mirroring the workflow of bioinformatics experts. The dataset contains 5,600 high-quality samples covering five core tasks across three stages: Screening, Quality Control, and Phenotype Annotation. Our evaluation of eight LLMs reveals that general-purpose reasoning models significantly outperform random baselines in phage contig identification and host prediction, demonstrating promising potential for genomic understanding. However, they exhibit significant limitations in complex reasoning tasks involving long-range dependencies and fine-grained functional localization. These findings highlight the necessity of developing next-generation models with enhanced reasoning capabilities for biological sequences.