Year: 2026
Type: preprint
Abstract: Transformer-based genomic sequence models represent an emerging frontier in computational biology. Yet, their embeddings have not yet shown the same level of predictive power as natural and protein language models, indicating a gap between current implementations and theoretical promise. Existing benchmarks for DNA language models primarily focus on classifying regulatory elements in eukaryotic genomes, leaving open the fundamental question of whether these models learn sequence-level features across whole genomes. We introduce LAMBDA, a benchmark designed to rigorously evaluate genome language model embeddings through phage-bacteria sequence discrimination across four categories of increasing complexity: probing tasks, fine-tuning assessments, diagnostic tests, and genome-wide prophage detection. Our comprehensive analysis of current genomic language models provides novel insights into the importance of training data quality relative to model size, the need for domain-specific training, and the application of genomic language models for detecting prophage sequences. This benchmark represents a challenging genomic annotation task in the bacterial domain and addresses a key computational problem with direct relevance to microbiology and medicine.
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
Funders: National Science Foundation, University of Utah, National Centre for Supercomputing Applications, National Institutes of Health, U.S. National Library of Medicine, Office of Advanced Cyberinfrastructure, National Center for Advancing Translational Sciences, NHLBI Division of Intramural Research
Awards: G1353381076, G1453181110, G1503004319, Track 4: Advanced CI Coordination Ecosystem: Monitoring and Measurement Services, G298330452, G34258105, G3562386289, G360632185, Track 1: ACCESS Resource Allocations Marketplace and Platform Services (RAMPS), G4010157413, G4275568664, URoL:EN: Understanding the rule of life facilitating the proliferation of toxic cyanobacterial benthic mats in flowing freshwaters, G4403015908, ACO: An Open CI Ecosystem to Advance Scientific Discovery (OpenCI), Track 3: COre National Ecosystem for CyberinfrasTructure (CONECT), G5156331529, G5367127336, G5464919783, G7526614437, G8108599699, Track 2: Customized Multi-tier Assistance, Training, and Computational Help (MATCH) for End User ACCESS to CI, G8499740153, G9000103499, G9000162286, CTSA UM1 Program at University of Utah
OpenAlex ID: W7141556418
Cites:
Cited by:
Related to: