The abstract is read at the publisher; the summary is JClub's.
PhageLysClass identifies phage lytic proteins and distinguishes Endolysins, Depolymerases, and virion-associated lytic enzymes, achieving about 0.89 MCC under strict R30/S30 conditions.
- Why it matters: Uncertain annotations, modular architectures, and class imbalance limit reliable sequence-based classification of phage lytic proteins, especially when evidence is incomplete or labels are missing.
- What they did: Using 442,329 unique phage sequences with 11,867 evidence-supported positives, the framework ran 7 supervised and 2 positive-unlabeled tasks across redundancy and train-test homology settings, testing ESM2-T30 representations.
- The result: ESM2-T30 consistently performed best, and the hierarchical classifier produced ~0.89 MCC in a pre-refitting benchmark; applied without retraining to 77,897 sequences, it flagged Depolymerase-like candidates and separated reference-supported from extrapolated assignments.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.