These notes are built from the UniProt record, the cached GOA rows and the cached
publications listed below; the Falcon deep-research report arrived after the review
was written and is cross-checked at the end of this file.
UniProt records the problem explicitly for both paralogues: "There is some
confusion regarding the nomenclature of this gene. In the literature, Tlr11 is
frequently referred to as Tlr12 and vice-versa"
(file:mouse/Tlr11/Tlr11-uniprot.txt). MGI records the same for the knockout
allele used in most of the literature (Tlr12
exists regarding the identity of Tlr11 and Tlr12. All of the references listed
below identify this allele as an allele of Tlr11. However, all of the associated
sequence information indicates that this is, in fact, an allele of Tlr12."
The direction of the inversion can be settled from sequence data in the cached
full texts, independently of what the papers call the proteins:
MPRMERHQFC SVLLILILLT ..., has RefSeqMGRYWLLPGL LLSLPLVTGW ..., RefSeq NP_991392,Koblansky et al. cloned the receptor they call TLR12 as
PMID:23246311
— i.e. the entry MGI calls Tlr11 — and note its construct "included the
signal sequence found in the first 30 amino acids", matching Q6R5P0's 1..30
signal. Raetz et al. give both cloning primers:
PMID:24078692 for their "TLR11" (reading frame
ATG GGC CGC TAC TGG CT = M-G-R-Y-W, the Q6QNU9 N-terminus) and
PMID:24078692 for their "TLR12"
(ATG CCC CGC ATG GAG CG = M-P-R-M-E, the Q6R5P0 N-terminus). Pifer et al. cloned
their "Tlr11" with PMID:21097503, again the
Q6QNU9 N-terminus.
Conclusion: literature "TLR12" (Koblansky 2013, Raetz 2013) = Q6R5P0 = MGI
Tlr11 = this gene; literature "TLR11" (Zhang 2004, Yarovinsky 2005, Pifer 2011,
Mathur 2013) = Q6QNU9 = MGI Tlr12. UniProt's own reference sets agree: PMID:15001781
(uropathogenic bacteria) is cited under Q6QNU9 with FUNCTION and DISRUPTION
PHENOTYPE, not under Q6R5P0.
Compared Tlr11-deep-research-falcon.md against the finished review. The report
is written around the literature name, not the database symbol. Its subject is the
receptor the literature calls TLR11, i.e. Q6QNU9 / MGI Tlr12, so most of its content
(Hatai's flagellin and profilin binding assays, the MIC3 candidate ligand, the testis
expression survey) belongs to the paralogue's review and was cross-filed there. The
report never states the inversion; it only remarks that the gene "was historically
also referred to as" the other name.
Agreement, for the parts that are about this entry (the report's "TLR12" statements
and the family-level material): intracellular endosomal/endolysosomal rather than
surface localisation; UNC93B1 dependence; direct profilin binding by the ectodomain
at endolysosomal pH; obligate heterodimer with the paralogue in conventional
dendritic cells and macrophages; MyD88-dependent signalling with the paralogue
supplying MyD88 recruitment; IL-12 as the defining output; absence of a functional
human counterpart. These all match decisions already in the review.
Additions taken up, each verified in a primary paper:
description corrected; the human absence, which is the point that matters for thedescription, the GO:0002224 summary and the core-function description no longerNot taken up: the report's claims about plasmacytoid-dendritic-cell homodimer function
and about this receptor being more essential than its paralogue are Koblansky findings
stated under the inverted name, and the review already records the homodimer capability;
no annotation turns on them. The MIC3 candidate ligand and the testis expression are
paralogue-side and in any case too preliminary for a GO annotation.
Actions: unchanged for every existing annotation and for the proposed GO:0042832. Changes
were confined to description, two narrative fields and one new suggested question. The
Falcon file is deliberately not cited as supported_by on any annotation, because its
gene-name mapping is inverted and a reader following the citation would be misled; the
validator warning about unused deep research is accepted for that reason.
GOA refreshed via just fetch-gene mouse Tlr11: unmodified, all 8 GOA rows already covered; the 9th review row (GO:0042832 defense response to protozoan) is a proposed NEW annotation not in GOA, as expected. Status already COMPLETE.
Ortholog consistency: human TLR11 is a pseudogene and TLR12/TLR13 are absent from the human genome, so there are no human ortholog reviews to reconcile; grep -ril 'Toll-like receptor 1[123]' genes --include=*-ai-review.yaml returns only the three mouse reviews. The receptor-pair biology is therefore reviewed on its own evidence.
Family consistency (vs Tlr12, Tlr13 and human endosomal TLRs TLR3/TLR7/TLR8/TLR9): consistent. GO:0002224 IBA ACCEPT, GO:0038023 IBA -> MODIFY to GO:0038187 pattern recognition receptor activity, and GO:0038187 as the core molecular_function all match the family convention (human TLR8 performs the same GO:0038023->GO:0038187 MODIFY; every family core MF is GO:0038187). The plasma-membrane IBA is handled by MODIFY to GO:0005768 endosome here, where the human endosomal TLRs instead KEEP_AS_NON_CORE/MARK_AS_OVER_ANNOTATED because they carry separate experimental endosome-membrane rows; the mouse entry has no such separate row, so MODIFY is the correct route and this is a justified difference, not an inconsistency.
Heterodimer: Tlr11-Tlr12 heterodimer treatment is consistent with Tlr12 PMID:23246311. Both reviews describe the obligate heterodimer in prose and core_functions, assert no complex cellular-component term, and record that the paralogue supplies MyD88 recruitment. The nomenclature inversion (literature TLR12 = this entry Q6R5P0) is handled identically in both files, with the flagellin/UPEC rows left UNDECIDED here and proposed NEW on Tlr12 per the sequence evidence [PMID:24078692, MGI allele note on MGI:3045751]. No defects; outcome no_change.