DMS3-25

Summary: CACAO row, and the only GO annotation this protein has. GOA actually carries two rows for this exact tuple of term, evidence code, reference and qualifier, differing only in ECO subtype (ECO:0001175 deletion mutation phenotypic evidence, ECO:0007089 loss-of-function mutant phenotype evidence); the review schema does not represent ECO, so this one entry covers both. Together they say the curator saw a deletion and a loss-of-function allele of this gene lose antiphage defence. The cited paper describes Rip1, a 164-residue prophage-encoded protein that senses the oligomeric portal or small terminase ring of an infecting phage, polymerises on it, and assembles into a membrane-disrupting pore that blocks virion assembly and kills the host cell early. Its abstract is the only readable text (Nature paywall, no PMC or Europe PMC record, no preprint) and does not name DMS3, and it describes two kinds of protein on opposite sides of the same defence: Rip1, which executes it, and the phage rings, which trigger it. Which one this entry is was therefore settled by sequence, using the two chains of the paper's own cryo-EM structure (PDB 9OOX, HK446 Rip1 bound to phage T6 small terminase) as the two reference points. A0SMN5 aligns to Rip1 over its full length at 30.8% identity with the C-terminal twin CxxC module conserved, and scores negative against the small terminase; its gene is also divergent from the DMS3 packaging operon rather than in it. This protein is the defence effector, so GO:0051607 defense response to virus is the right term, and it is the term GO uses for bacterial antiphage defence systems.

Reason: The claim is correct, supported by mutant-phenotype experiments on this gene in the cited paper, and GO:0051607 is the appropriate granularity: GO has no abortive-infection term, and none of this term's non-regulation children fits a bacterial pore-forming defence (GO:0009616 is RNAi-mediated, GO:0140374 antiviral innate immune response is the metazoan innate-immunity branch, and GO:0044790 suppression of viral release by host names release rather than assembly). The only defect is the GAF qualifier. acts_upstream_of_or_within_positive_effect describes a gene product that influences a process from outside it, which is exactly what the activating phage portal and terminase rings do, and is not what this protein does: it integrates sensing and effector activity and forms the pore itself, so involved_in is correct. That is a source-record field error, recorded in source_defects, and not a reason to reject the claim; the same resolution was reached for the equivalent row in genes/9CAUD/acrIIA5. Retained as the core function of the gene.

Summary: GOA gives this protein no molecular function at all. The cited paper's mechanism for the family is a pore: the protein polymerises on the phage assembly ring and inserts into the membrane, disrupting it. The sequence analysis in this review's bioinformatics folder places DMS3-25 in that family, aligning it over its full length at 30.8% identity to HK446 Rip1 (UniProtKB:K7P861, the 164-residue protein of PDB 9OOX chain L, which the paper characterises), with the C-terminal twin CxxC module conserved and no detectable similarity to the activating small terminase. GO:0140911 pore-forming activity is the GO term for this activity; its synonyms are holin and pinholin activity, so the term is in use for proteins that permeabilise the membrane of the cell expressing them, as the Rip1 family and the gasdermins do, even though its definition as written says "the membrane of another cell".

Reason: Added to give the gene the molecular function that its one existing annotation, a biological process, leaves unstated, and to make explicit the activity asserted in core_functions. The inference is a molecular one about what this polypeptide does, not a process role, and it rests on full-length homology to a protein whose pore-forming activity is experimentally established, so ISS with UniProtKB:K7P861 as the source is the honest evidence code. It is deliberately not IDA or IMP: no pore has been demonstrated for the DMS3 protein itself, which is recorded as a knowledge gap. The definitional mismatch in GO:0140911 is raised in suggested_questions rather than treated as a reason to withhold the term.