Illuminating phage dark matter beyond the reach of structural alignments with PhageFlare
- 1Hong Kong Baptist University
Abstract
Many phage proteins remain functionally uncharacterized despite advances in structural-similarity-based annotation. We introduce PhageFlare, a deep learning framework for predicting phage protein functions lacking significant structural matches to annotated references. Across three structure-disjoint benchmarks, where no significant Foldseek match was detected between training and test proteins (E-value thresholds of 10-3, 10-4, and 10-2), PhageFlare achieved macro-averaged F1 scores of 0.871, 0.889, and 0.738 outperforming the strongest baseline (0.789, 0.814, and 0.732). Systematic evaluations demonstrated that protein language model selection critically influences performance, with ESM-C achieving stronger overall performance than ESM-2 and ProtBert. Embedding-guided case studies revealed local structural correspondences among proteins with limited global similarity. Applied to the efam and Metagenomic Gut Virus datasets, PhageFlare annotated 38.0% and 51.8% of proteins unresolved by Foldseek, respectively. Furthermore, PhageFlare identified lysogenic potential in 1,866 prophage-designated genomes misclassified as lytic by BACPHLIP. These results show that PhageFlare complements structural-similarity-based annotation and expands functional discovery across viral dark matter.
Related
- Illuminating Antarctic Microbial Dark Matter: In situ Cultivation Reveals Rare Taxa with Broad Resistome and Virulence Phenotypes.
- Bramble: projection of spliced genomic alignments into transcriptomic space for improved transcript quantification
- Illuminating gendered ageism with the WHO ageism experiences scale: a psychometric analysis and gender comparisons in a multi-country dataset.
- Beyond the uniform: uncovering structural complexity within Adélie penguin colonies
- PhyloScore: Improved covariation score for RNA alignments using phylogeny
BibTeX
@misc{wang2026illuminating,
title = {{Illuminating phage dark matter beyond the reach of structural alignments with PhageFlare}},
author = {C. Wang and Y. Xu and Z. Li and L. Zhang and L. Zhang},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.28.762509},
url = {https://www.biorxiv.org/content/10.64898/2026.10.28.762509v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Illuminating phage dark matter beyond the reach of structural alignments with PhageFlare
AU - C. Wang
AU - Y. Xu
AU - Z. Li
AU - L. Zhang
AU - L. Zhang
PY - 2026
DA - 2026/10/28
PB - bioRxiv
DO - 10.64898/2026.10.28.762509
UR - https://www.biorxiv.org/content/10.64898/2026.10.28.762509v1
AB - Many phage proteins remain functionally uncharacterized despite advances in structural-similarity-based annotation. We introduce PhageFlare, a deep learning framework for predicting phage protein functions lacking significant structural matches to annotated references. Across three structure-disjoint benchmarks, where no significant Foldseek match was detected between training and test proteins (E-value thresholds of 10-3, 10-4, and 10-2), PhageFlare achieved macro-averaged F1 scores of 0.871, 0.889, and 0.738 outperforming the strongest baseline (0.789, 0.814, and 0.732). Systematic evaluations demonstrated that protein language model selection critically influences performance, with ESM-C achieving stronger overall performance than ESM-2 and ProtBert. Embedding-guided case studies revealed local structural correspondences among proteins with limited global similarity. Applied to the efam and Metagenomic Gut Virus datasets, PhageFlare annotated 38.0% and 51.8% of proteins unresolved by Foldseek, respectively. Furthermore, PhageFlare identified lysogenic potential in 1,866 prophage-designated genomes misclassified as lytic by BACPHLIP. These results show that PhageFlare complements structural-similarity-based annotation and expands functional discovery across viral dark matter.
ER -