K9IWC0 Research Notes

Key findings

2026-07-31 compliance review

This is NPPC (CNP), not NPPB (BNP). The UniProt name is a paralogue
misassignment. Three independent lines say so, and they were all sitting in
plain view:

  1. Internal contradiction in the UniProt record itself. The ARBA-derived
    name is [file:DESRO/K9IWC0/K9IWC0-uniprot.txt "RecName: Full=Natriuretic
    peptides B"], but the PANTHER subfamily in the same record is
    [file:DESRO/K9IWC0/K9IWC0-uniprot.txt "PANTHER; PTHR12167:SF2; C-TYPE
    NATRIURETIC PEPTIDE; 1."].
  2. The source publication names it CNP. The EMBL record behind this
    accession (JAA44981.1, TISSUE=Salivary gland) comes from the Vampirome study,
    now cached as PMID:23411029, which has a section headed "C-type natriuretic
    peptide (CNP)" and states PMID:23411029.
  3. Sequence. Ran a reproducible check in K9IWC0-bioinformatics/
    (classify_natriuretic_peptide.py; sequences fetched from the UniProt REST
    API at run time, alignment by Biopython, nothing hardcoded):
    72.3% identity to human NPPC vs 31.9% to each of NPPB and NPPA, and the
    precursor terminates exactly at the second ring cysteine — the CNP-diagnostic
    pattern, since NPPA and NPPB precursors carry 5 and 6 trailing residues
    respectively. Mature ring CFGQKLDRIGALSGLGC vs human CNP
    CFGLKLDRIGSMSGLGC: 3 differences out of 17. Against the BNP ring: 8 of 17.

Why it matters for annotation. BNP signals through NPR1/GC-A; CNP signals
through NPR2/GC-B. The UniProt FUNCTION sentence [file:DESRO/K9IWC0/K9IWC0-uniprot.txt
"Acts by specifically binding and stimulating NPR1 to"] is therefore BNP biology
attached to the wrong protein, and any NPR1-specific inference drawn from it
would be wrong. The Vampirome paper gets it right: PMID:23411029.
Recorded as a CURATION knowledge gap to be reported upstream to UniProt.

The deep-research file inherited the error wholesale — it opens with
[file:DESRO/K9IWC0/K9IWC0-deep-research-falcon.md "UniProt K9IWC0 corresponds to
a natriuretic peptide B precursor (BNP prohormone) annotated in Desmodus
rotundus (vampire bat)"] and then spends most of its length on human NT-proBNP
heart-failure diagnostic thresholds, NPR1 agonist antibodies, and AF-adjusted
cut-offs — none of which has anything to do with a salivary vasodilator peptide.
Marked relevance: LOW.

Annotation change. GO:0097746 blood vessel diameter maintenance had been
marked over-annotated. It is direction-neutral (covers constriction and
dilation), and the direction is known: PMID:23411029. Changed to MODIFY → GO:0042311 vasodilation
(verified via OLS as a direct child of GO:0097746). The other four annotations
were sound; GO:0006182 and GO:0007168 are ACCEPTed with the receptor
identity corrected in the summary, since neither GO term names a receptor.