human glycosylation annotations over-inflated

[ValWood]

A lot of these appear to be "glycosylated" not involved in glycosylation

first one I looked at: a cholesterol transporter https://www.uniprot.org/uniprot/O15118 IDA https://www.uniprot.org/citations/10821832

[ValWood]

https://www.uniprot.org/uniprot/P07306 TAS via reactome (receptor for exocytosed glycoprotein) https://www.uniprot.org/uniprot/P07911 TAS via reactome

P98088 MUC5A_HUMAN Mucin-5AC O-glycan processing Source: Reactome Q02505 MUC3A_HUMAN Mucin-3A O-glycan processing Source: Reactome Q02817 MUC2_HUMAN Mucin-2 O-glycan processing Source: Reactome Q5SSG8 MUC21_HUMAN Mucin-21 O-glycan processing Source: Reactome Q8N307 MUC20_HUMAN Mucin-20 Q8N387 MUC15_HUMAN Mucin-15 Q8TAX7 MUC7_HUMAN Mucin-7 Q8WXI7 MUC16_HUMAN Mucin-16 Q96DR8 MUCL1_HUMAN Mucin-like protein 1 Q99102 MUC4_HUMAN Mucin-4 Q9HC84 MUC5B_HUMAN Mucin-5B Q9UKN1 MUC12_HUMAN Mucin-12 Q9H195 MUC3B_HUMAN Mucin-3B Q9H3R2 MUC13_HUMAN Mucin-13 Q685J3 MUC17_HUMAN Mucin-17 Q6W4X9 MUC6_HUMAN Mucin-6 Q7Z5P9 MUC19_HUMAN Mucin-19 P27918 PROP_HUMAN Properdin protein O-linked fucosylation Source: Reactome

[deustp01]

Those TAS from us should go away with our next release at the end of the month.

[ValWood]

Brilliant. I'll close this one. I noticed some others which might be slightly different reasons. so I will open tickets for these later......But am I OK to assume that any resulting in "target of process annotated to process" are fixed for all processes?

v

[deustp01]

"OK to assume that any resulting in "target of process annotated to process" are fixed for all processes?"

The discrepancy arises at the level of individual reactions (molecular functions) and propagates to the pathways (processes) that those reactions are parts of. Where the reaction involves catalytic activity (GO:0003824) and its children or transporter activity (GO:0005215) and its children, the discrepancies should be fixed. In the cases of proteins and complexes traversing the nuclear pore or proteins associated with intraqcellular vesicles moving to the cell membrane and extracellular space, the fixes will take longer.

[ValWood]

reopening as there are non -reatcome ones above.

[ValWood]

From MGI @ukemi @hdrabkin please assign appropriate person

• [ ] 1. Q9BQQ3 GORS1_HUMAN Golgi reassembly-stacking protein 1 These findings substantiate the role of GRASP65 in continuity of the cis Golgi network required for proper glycosylation, while showing that neither this continuity nor GRASP65 itself are essential for the viability of a complex organism.

ISS from mouse http://www.uniprot.org/citations/24795147 Seems to be a phenotype due to Golgi disruption rather than an role in glycosylation

• [ ] 2. Q15643 TRIPB_HUMAN Thyroid receptor-interacting protei.. also seems to be from Golgi abnormalities via UniProtKB:E9Q512 (mouse) https://www.ncbi.nlm.nih.gov/pubmed/20089971?dopt=Abstract

• [ ] 3. Q86UW2 OSTB_HUMAN Organic solute transporter subunit ... via mouse UniProtKB:Q80WK2 http://europepmc.org/articles/PMC3375545 appears to be glycosylated Formation of the transporter complex is also coupled to post-translational modifications of Ostα; i.e. stable interaction between the subunits is required for progression through the biosynthetic-secretory pathway and to generate a glycosylated form of Ostα (5, 9, 10). Thus, the glycosylation status of Ostα can be utilized as an index of interaction between the two subunits.

[ValWood]

UniProt @ggeorghiou

• [x] glycosylated P60604 UB2G2_HUMAN Ubiquitin-conjugating enzyme E2 G2 P55072 TERA_HUMAN Transitional endoplasmic reticulum ... Q86TM6 SYVN1_HUMAN E3 ubiquitin-protein ligase synovio...
  protein N-linked glycosylation via asparagine IMP, PMID:22607976 appears to be glycosylated

• [x] glycosylated Q12893 TM115_HUMAN Transmembrane protein 115
  protein glycosylation PMID:24806965 appears to be glycosylated (uniprot function line also needs updating)

[ukemi]

@vanaukenk and @pgaudet This is a nice edge case for the new qualifiers and might be worth discussing on a call. You need a good Golgi apparatus to get proper glycosylation, but if we make the cut off of the glycosylation being the enzymatic steps, then the organization of the Golgi would be upstream of it. In my opinion it is still very important to capture this kind of information in some way. The last case is even more interesting. The complex between the beta subunit and the alpha subunit is required for the alpha subunit to be glycosylated. The beta is annotated to positive regulation, but if we make the cut off being the enzymatic steps, it would act upstream as well.

[ggeorghiou]

All of these annotations have been deleted or updated accordingly. Closing this ticket.