SMIM3 (NID67, C5orf62) notes

2026-10-08: initial review (MICROPROTEINS Tier 2)

Identity. UniProt Q9BZL3, 60 aa, Swiss-Prot reviewed, MANE-select canonical gene product on
5q33.1 (not an alternative ORF, so the standard genes/human/SMIM3/ folder applies). One
predicted TM helix (residues 20-40); a "Cleavage" site 19-20 is propagated by similarity from
the mouse entry (Q99PE6) and is not experimentally shown for human.

Literature. PubMed SMIM3[tiab] OR NID67[tiab] returns 10 papers. Only two say anything
about the protein:

GOA (60 rows).
- 58 x GO:0005515 protein binding IPI from four binary-interactome screens (HuRI 47, HI-II-14 9,
Sahni 2015 2, CCSB-HI1 1). Partners are ~55 mostly unrelated membrane proteins (PMP22, MAL,
BSCL2, CNIH1/3, EMP1/3, claudins, tetraspanins, ATP6V0C, SLC22A1, CLEC7A...): the typical
sticky single-TM-helix Y2H signature also seen for STRIT1. All REMOVE per project policy.
- 1 x GO:0042802 identical protein binding (HuRI self-pair): MARK_AS_OVER_ANNOTATED.
- 1 x GO:0016020 membrane IEA (SubCell): ACCEPT; the only supportable statement.

Outcome. No MF or BP term is supported. Core function records membrane location only. No
NEW proposals: the AML proliferation phenotype is a knockdown necessity result with no
mechanism, and fails the participation test.