Applied review and annotation-reviewer instructions. Verified current UniProt accession, PAO1 organism (NCBITaxon:208964) and gene symbol before normal just fetch-gene. Preserved every source annotation field, including evidence, qualifier, supporting entities and reference; only review decisions were added. No GO-CAM index entry was found for this accession. All PMID quotes refer to the normal publication cache; no cached source, GOA or UniProt file was hand-edited. Abstract-only sources were not treated as complete papers. No unsupported NEW process rows were added.
The actual default-provider attempts just deep-research-falcon PSEAE alg8 --fallback perplexity-lite failed: Edison/Falcon HTTP403 and fallback HTTP401 insufficient quota. These notes record manual primary-source research and are not a provider report. just fetch-fitness PSEAE alg8 could not retrieve a local/remote FEBA result and did not establish an organism mapping/result; no fitness phenotype is claimed. Source and provider attempts were performed alongside publication caching. UniProt family/domain memberships are classification evidence, not DE/GO/ARBA predictions used as proof of chemistry.
Alg8 is the catalytic GDP-mannuronate polymerase. PMID16391057 assays cell-envelope fractions, not a purified isolated enzyme. PMID18524915 provides membrane topology and catalytic-motif mutagenesis. The two hyaluronan IBAs at PTN000530809 cross the demonstrated substrate specificity and are removed; the target in WITH/FROM is not circular evidence. Broad polysaccharide synthesis is refined; ECM assembly is corrected laterally to actual polymer biosynthesis, not claimed as a descendant refinement.
UniProt DR:PANTHER PTHR22913:SF12 has official name MANNURONAN SYNTHASE; the broader PTHR22913 label HYALURONAN SYNTHASE does not establish hyaluronan specificity for this member. PF13641 Glyco_tranf_2_3 is the GT-domain support.
The prior lateral MODIFY of GO:0085029 is superseded by KEEP_AS_NON_CORE after inspecting full PMID:25968647 (Alginate Polymerization and Modification Are Linked in Pseudomonas aeruginosa), Fig. 9 and Table 2. Native PDO300 alg8 deletion and complementation were compared in continuous-culture flow-cell biofilms: "The PDO300Δalg8(pBBR1MCS-5:alg8) and PDO300Δalg44(pBBR1MCS-5:alg44) strains established heterogeneous highly structured biofilms (Fig. 9)." The deletion control lacked elevated structured architecture. Alg8 performs the polymerization that constructs the matrix component, so this is a participation argument supported by target-specific architecture measurements, not necessity evidence alone. Matrix assembly is a conditional non-core role; alginate polymerization remains core. The term definition concerns aggregation, arrangement and bonding of extracellular matrix. This supports retention more directly than the related-strain mt-2 matrix evidence used for PSEPK; no equivalent native architecture assay was found for AZOVI, whose over-annotation judgment is unchanged. This is an existing IBA review, not a NEW process annotation. PAINT ancestry remains unresolved; no failed node placement is asserted. All original annotation fields are unchanged.
Hyaluronan-specific propagation is treated consistently as REMOVE across the three Alg8 reviews; supported alginate roles remain independently represented. Live UniProt PANTHER cross-references show that SF12 itself contains S. pyogenes HasA P0C0H0 (hyaluronan chemistry; PMID:8276791) and NodC P04340/P26024, in addition to the three Alg8 proteins. The SF12 label MANNURONAN SYNTHASE is therefore not a substrate-specific selector. GO:0085029 was checked live: matrix aggregation/arrangement/bonding, with no animal restriction. Retained the evidence gradient: Pa target-specific architecture, KT2440 related-strain inference, Azotobacter over-annotation without a comparable matrix-architecture result. Substituting a biofilm term would change the biological claim rather than simply refine it.
Audit only; the existing review was not rewritten. Seed paper PMID:18524915 is
abstract-only in the cache and the PubMed MCP did not return PMC full text.
The abstract shows target-specific catalytic evidence, not just a phenotype:
PMID:18524915;
PMID:18524915.
Seeds of PTN000530809 (.cache/paint/IBD.gaf, column 8): GO:0000271 is seeded by
MGI:109599, MGI:107821, MGI:106590 (mouse Has1, Has2, Has3) and Q52463 (Alg8). The
hyaluronan terms (GO:0050501, GO:0030213) and GO:0085029 at the same node are seeded
only by vertebrate HAS proteins.
| PTN node | node term | n_seeds | leaf IBAs | seed annotation (term, evidence, PMID) | verdict | rationale | other seeds | IBD stands without this seed? |
|---|---|---|---|---|---|---|---|---|
| PTN000530809 | GO:0000271 polysaccharide biosynthetic process | 4 | 85 (31 taxa) | GO:0042121 alginic acid biosynthetic process, IMP, PMID:18524915 (PseudoCAP) | SOUND | Active-site alanine substitutions abolish polymerization in vivo, so Alg8 itself makes the polymer. The broad node term suits a glycosyltransferase-2 clade that makes several polysaccharides. | Mouse Has1, Has2, Has3 (hyaluronan synthases) | Yes. Hyaluronan synthesis is polysaccharide biosynthesis. No GO_CENTRAL defect on this term. |
Node-level problem found during the audit (not on the seeded term): the hyaluronan IBDs at
the same node are placed too high. Leaves of PTN000530809 carrying the GO:0050501 IBA
(from .cache/paint/leaf.gaf.gz) include, besides vertebrate HAS genes:
O67594 alg, Q9FBM8 SCO5051, Q52463 alg8, Q97YM4 nodC-like, P26024 nodC
and fungal (AFUA_8G02320, SNOG_10021, SS1G_06158) and Entamoeba (EHI_044840, EHI_170480)
proteins. NodC makes chitooligosaccharides and Alg8 makes mannuronan, so the hyaluronan
terms do not suit this clade. Recorded once, as a GO_CENTRAL CHECK defect (PROBABLE) on the
GO:0050501 IBA row; it covers the GO:0030213 row too. The existing REMOVE actions on both
hyaluronan rows are unchanged.
Not changed: the GO:0000271 IBA row is MODIFY to GO:0042121; under the later tier-2
convention it would be TOO_GENERAL (GO:0042121 is_a GO:0000271 and is accepted here). The
claim is the same either way, so it was left as is.