fimT

Summary: UniProt subcellular-location mapping to plasma membrane; the UniProt location (cell inner membrane, single-pass) is itself an ARBA prediction.

Reason: The ARBA location is not evidence on its own, but FimT carries the N-methylation (class III prepilin) motif and N-terminal hydrophobic helix of type IV prepilins, which reside in the inner membrane before PilD processing and assembly, and it belongs to the GspH/FimT minor pilin family whose members are inner-membrane targeted. Inner membrane is where a prepilin resides and is assembled from.

Summary: InterPro2GO mapping from IPR022346 (T2SS GspH).

Reason: IPR022346 covers the whole GspH/FimT family, which includes the type IV pilus minor pilins FimT and FimU as well as the T2SS pseudopilin GspH. FimT is encoded in the type IV pilus gene region and its only known in vivo role is in fimbrial biogenesis; P. aeruginosa has its own Xcp T2SS GspH, XcpU (PA3100, Q00515). There is no evidence that FimT is a T2SS component. The domain-to-complex mapping is wrong for this family member, as in the fimU review.

Summary: InterPro2GO mapping from IPR022346 (T2SS GspH).

Reason: Same family-level over-reach as the complex row. No study links FimT to Xcp-dependent secretion; the experimental data place it in type IV pilus biogenesis and, in vitro, DNA binding. The Xcp system has its own GspH (XcpU).

Summary: PseudoCAP IDA (2019-03-11) citing Chiang et al. 2008, an ATPase study of purified PilB, PilT and PilU. The term is the same off-by-one id slip found on the pilU row from that paper, and the paper's abstract concerns only the three NTPases, not FimT.

Reason: The claim is wrong whatever the full text contains. A live QuickGO query (2026-11-03) shows four PseudoCAP rows citing PMID:18174131, all dated 2019-03-11: pilB (P22608) and pilT (P24559) to GO:0016887 ATP hydrolysis activity, and pilU (G3XCX3) and fimT (Q9HVN0) to the adjacent id GO:0016888. The pilU review established that GO:0016888 was keyed for GO:0016887. FimT is a GspH-family minor pilin with no nuclease domain, and no study reports nuclease activity for any FimT. Repointing to GO:0016887 would also be wrong: the abstract names PilB, PilT and PilU as the purified and mutated motor proteins, and FimT is not an NTPase. The fimT row is most likely a copy of the pilU row attached to the wrong gene product, so it should be deleted rather than corrected. This is a demonstrable keying error, not a second-guessed experiment; the full text was not available (abstract only), which is why the gene-product half of the defect is PROBABLE.

Summary: PseudoCAP IDA from Alm and Mattick 1996: the fimT mutant is wild type, but FimT overexpression replaces FimU in fimbrial biogenesis.

Reason: FimT is a prepilin-like minor pilin of the GspH/FimT family whose paralogue FimU promotes type IV pilus assembly, and overexpressed FimT can take over FimU's role, so FimT can itself do the minor-pilin work of priming assembly. Its contribution in wild-type cells is subtle (the single mutant has no phenotype), which the paper says. The evidence described in the abstract is genetic (mutant and overexpression suppression), which fits IGI or IMP better than IDA; I cannot read the full text, so this is noted rather than filed as a defect.

Summary: Purified P. aeruginosa FimT binds a 30 bp dsDNA in EMSA, with an affinity of the same order as Legionella FimT; P. aeruginosa FimU does not bind.

Reason: Direct in vitro binding assay on the purified P. aeruginosa protein, with FimU as an internal negative comparator. The term is as specific as the assay supports (30 bp dsDNA; ssDNA was tested only for the Legionella protein, and binding there is non-sequence-specific). The physiological significance in P. aeruginosa, which is not known to be naturally transformable, is unknown, so no process term is proposed.