Summary: InterPro2GO transfer of Dol-P-Man synthase activity from the DPM1-like family signature (IPR039528) to the 51-residue Y66H1A.2b product. The activity is right for the dpm-1 gene, whose full-length product Q9TYJ7 sits in the DPM1 subfamily PTHR43398:SF1 with human DPM1 and yeast Dpm1. It is not right for this product, which lacks the catalytic domain.

Reason: The catalytic activity of DPM1-family synthases lives in the GT-A glycosyltransferase fold. In the canonical worm product Q9TYJ7 this domain spans residues 8-176. U4PF58 is identical to canonical residues 189-239 only, so none of that domain is present. Mapped onto human DPM1 (64.7% identity to Q9TYJ7 over 238 aligned columns), the segment covers residues 211-260. It contains 2 of the 13 GDP-mannose binding residues annotated by similarity to the archaeal synthase structure, and none of the other 11 GDP-mannose binding residues (human positions 32-147), nor the Mg2+/Mn2+ binding residue at human position 120. In mammals the catalytic subunit also has to be tethered and stabilised by DPM3. An isolated C-terminal tail without the fold cannot catalyse the transfer from GDP-mannose to dolichyl phosphate. No worm-specific experiment on either dpm-1 product was found. The family signature matched on sequence identity to the gene's tail, not on a retained catalytic domain, so the electronic mapping fails for this product. This REMOVE concerns only the U4PF58 product. The same term is expected and appropriate for the full-length product Q9TYJ7, as noted in the source_defects entry.