Glycobiology pathway modules
Index of the modules/ entries that belong to the
Glycobiology project. The 17 core modules below, plus
lysosomal_glycogen_degradation, were curated independently of the project's
seven exemplar gene reviews
and were linked from no project page at all; this page wires them in.
galactose_leloir_pathway and the non-animal modules in the last section already
belong to other projects and are listed only as cross-references.
The modules supply the axis the exemplar set lacks. The exemplars are
single-gene reviews chosen to probe annotation altitude and pleiotropy; the
modules are pathway-level ModuleReview documents that place glycogenes in
ordered biosynthetic and catabolic sequences. Together they reach glycoconjugate classes the exemplars never touch —
GPI anchors, glycosphingolipids, and the lysosomal catabolism of glycolipids and
glycosaminoglycans — rather than only the glycosyltransferase/lectin axes.
O-glycosylation remains a hole at module level: no module covers mucin-type
GALNT-initiated O-glycan biosynthesis, though the project's
Phase 3 gene batch
(GALNT1, C1GALT1, C1GALT1C1, GCNT1) now covers the initiation and first two
elongation steps as single-gene reviews. A module assembling them into the
ordered pathway is the obvious next step.
Scope note. A ModuleReview is not a GO-CAM. The decision to PUNT
GO-CAM/causal modelling in
GLYCOBIOLOGY-resource-reuse.md is unchanged by
this page; these modules are the repo's own pathway-decomposition format.
Core animal glycoconjugate modules
All are status: DRAFT. "Genes" counts the distinct gene reviews cited as
file: evidence by the module.
N-linked glycosylation
| Module | Genes | What it covers |
|---|---|---|
| n_glycan_llo_assembly_cytoplasmic | 6 | LLO assembly on the cytoplasmic ER face to Man5GlcNAc2-PP-dolichol (DPAGT1, ALG13/14, ALG1/2/11) |
| n_glycan_llo_assembly_lumenal | 6 | LLO completion on the lumenal face + glucosylation to Glc3Man9GlcNAc2-PP-dolichol (ALG3/9/12, ALG6/8/10) |
| oligosaccharyltransferase_complex | 9 | En-bloc transfer of the glycan to Asn; STT3A/STT3B catalytic subunits + accessory subunits (RPN1/2, DDOST, DAD1, OSTC, MAGT1, TUSC3) |
Donor and precursor supply
| Module | Genes | What it covers |
|---|---|---|
| dolichyl_phosphate_biosynthesis | 5 | The Dol-P lipid carrier itself (cis-PT/DHDDS-NUS1 → DOLPP1 → SRD5A3 → DOLK) |
| dolichol_phosphate_sugar_donor_supply | 7 | Dol-P-Man / Dol-P-Glc donor formation and flipping (DPM1/2/3, ALG5, MPDU1) |
| hexosamine_biosynthesis | 6 | Fru-6-P → UDP-GlcNAc, plus the GlcNAc salvage branch (GFPT1/2, GNPNAT1, PGM3, UAP1, NAGK) |
| sialic_acid_metabolism | 7 | Neu5Ac biosynthesis, CMP activation, Golgi import and catabolism (GNE, NANS, NANP, CMAS, SLC35A1, NEU1, NPL) |
| paps_sulfate_activation | 4 | PAPS synthesis and Golgi import — the donor for all glycosaminoglycan/glycolipid sulfation (PAPSS1/2, SLC35B2/3) |
GPI anchors
The five modules decompose GO:0006505/0006506 into its ordered stages, and
between them cover the PIG/PGAP gene family the project overview names.
| Module | Genes | Stage |
|---|---|---|
| gpi_anchor_glcnac_transferase | 6 | I — GPI-GnT complex (PIGA/C/H/P/Q/Y) |
| gpi_anchor_core_glycan_assembly | 6 | II — de-N-acetylation, inositol acylation, mannosylation (PIGL/W/M/X/V/B) |
| gpi_anchor_ethanolamine_phosphate | 4 | III — EtNP additions (PIGN/G/F/O) |
| gpi_anchor_transamidase | 5 | IV — transamidase complex, attachment to protein (PIGK/S/T/U, GPAA1) |
| gpi_anchor_remodeling | 3 | V — post-attachment lipid remodelling (PGAP1/2/3), i.e. GO:0120574 |
Glycolipids and glycan catabolism
| Module | Genes | What it covers |
|---|---|---|
| glycosphingolipid_biosynthesis | 10 | Ceramide → GlcCer/GalCer → LacCer → ganglio/globo/lacto series (UGCG, UGT8, B4GALT5/6, ST3GAL5, ST8SIA1, B4GALNT1, B3GALNT1, B3GNT5, A4GALT) |
| glycosphingolipid_lysosomal_degradation | 7 | The sphingolipidoses (ARSA, GALC, GBA, SMPD1, ASAH1 + the GM2A/PSAP activators) |
| heparan_sulfate_lysosomal_degradation | 7 | The mucopolysaccharidoses (IDS, IDUA, SGSH, HGSNAT, NAGLU, GNS, GUSB) |
| keratan_chondroitin_sulfate_lysosomal_degradation | 4 | Morquio / GM2 exolytic cascade (GALNS, GLB1, HEXA, HEXB) |
Adjacent
Glycan-metabolism modules at the project boundary — included for completeness,
not counted in the core cohort below.
| Module | Genes | Why adjacent |
|---|---|---|
| lysosomal_glycogen_degradation | 1 | Glycan catabolism, but of a storage polysaccharide rather than a glycoconjugate (GAA; Pompe) |
| galactose_leloir_pathway | 4 | Central carbon metabolism, but regenerates the UDP-galactose that galactosyltransferases spend (GALM, GALK1, GALT, GALE). Already curated under P_PUTIDA (ppu00052) |
Non-animal glycan modules (cross-reference)
These are already owned by other projects — listed here as cross-references
because they share the project's term landscape, not as glycobiology deliverables.
Bacterial and plant cell-surface glycan synthesis is where CAZy family coverage is
densest, so they are the natural test set for cazy2go even though they sit
outside the animal GO-usage audit.
| Module | Owning project | What it covers |
|---|---|---|
| adp_heptose_biosynthesis | P_PUTIDA (ppu00541) |
ADP-heptose for the LPS inner core (GmhA, HldE, GmhB, HldD) |
| dtdp_l_rhamnose_biosynthesis | P_PUTIDA (ppu00523/525) |
dTDP-L-rhamnose, an O-antigen/cell-wall sugar donor (rfbA/C/D, rffG) |
| peptidoglycan_precursor_biosynthesis | P_PUTIDA (ppu00550) |
Lipid II synthesis and export |
| peptidoglycan_recycling | P_PUTIDA (ppu00520) |
Pseudomonas-type anabolic recycling |
| glycogen_synthesis_and_mobilization | P_PUTIDA (ppu00500) |
Bacterial glycogen / α-glucan metabolism (glgA/B/P/X, galU) |
| cellulose_biosynthesis | PLANT_BIOENERGY | Plant cellulose synthase rosette; UDP-glucose → (1→4)-β-D-glucan |
The module gene cohort — a second, independent verdict baseline
The 17 core modules cite 100 distinct human gene reviews, all present in
genes/human/, with no overlap at all with the seven exemplars. That is a
substantially larger reviewed glycogene corpus than the project page previously
claimed, and it gives a second verdict distribution to compare against the
calibration set.
Counts below are computed from the existing_annotations[].review.action fields
of the 100 YAMLs (not transcribed from prose), over 2,735 annotations — every
one adjudicated, none left PENDING.
| Cohort | N | ACCEPT | NON_CORE | OVER | MODIFY | REMOVE | NEW |
|---|---|---|---|---|---|---|---|
| Module genes (100) | 2735 | 1729 (63.2%) | 486 (17.8%) | 422 (15.4%) | 73 (2.7%) | 13 (0.5%) | 12 (0.4%) |
| Exemplars (7) | 303 | 137 (45.2%) | 100 (33.0%) | 42 (13.9%) | 22 (7.3%) | 1 (0.3%) | 1 (0.3%) |
Three things the comparison shows:
- The project's central claim holds, and holds harder on the larger cohort.
REMOVE is 0.5% and 0.3% respectively; ~99.5% of annotations are retained in
some form. Mis-annotation really is altitude/specificity and pleiotropy rather
than wrong function, and that is now supported by 3,038 annotations rather
than 303. - The exemplar set is NON_CORE- and MODIFY-skewed by design, and the skew is
mostly one gene. LGALS3 alone contributes 62 of the exemplars' 100 NON_CORE.
The module cohort is dominated by narrow biosynthetic enzymes with little
pleiotropy, so it lands 18 points higher on ACCEPT. Neither distribution is
"the" glycogene baseline — the difference between them is a gene-class effect,
which is worth stating explicitly before the GOA closure query produces a third
number. - The module cohort is under-proposing new terms. 100 genes yielded 1
top-levelproposed_new_termsentry between them; the 7 exemplars yielded 8.
(This is a different field from theNEWcolumn above, which counts
annotations proposed against existing GO terms.) A 1-vs-8 split is unlikely to
be a real difference in GO coverage, and more likely reflects that the
module-driven reviews were not run with new-term proposal in view. It is a
concrete re-review target.
Caveat on status. 97 of the 100 carry status: INITIALIZED despite being
fully adjudicated (2 COMPLETE, 1 IN_PROGRESS). The status field is stale
rather than the reviews being incomplete, but it means the cohort cannot be
selected by status; it is selected here by module citation.