id: P76419
gene_symbol: yegV
locus_tag: b2100
taxon:
  id: NCBITaxon:83333
  label: Escherichia coli K-12
status: COMPLETE
description: >-
  YegV DeepECTF prediction review. The DeepECTF prediction of dehydro-2-deoxygluconokinase
  (EC 2.7.1.92) is a paralog overannotation error. YegV is correctly identified as a sugar
  kinase (first 3 EC digits correct) but the specific substrate assignment is wrong - that
  activity belongs to KdgK/b3526. YegV's substrate specificity remains unknown.
references:
  - id: PMID:37963869
    title: Functional annotation of enzyme-encoding genes using deep learning with transformer layers.
  - id: PMID:40703034
    title: Limitations of current machine learning models in predicting enzymatic functions for uncharacterized
      proteins.
predictions:
  - source_method: DeepECTF
    source_version: "2023"
    source_reference_id: PMID:37963869
    predicted_term:
      id: EC:2.7.1.92
      label: dehydro-2-deoxygluconokinase
    predicted_term_type: EC
    review:
      assessment: PLI
      confidence_score: 0
      error_type: PARALOG_OVERANNOTATION
      summary: >-
        Paralog incorrect. YegV is annotated as an uncharacterized sugar kinase (EC 2.7.1.-)
        in UniProt, so DeepECTF correctly predicted the first 3 digits of the EC number but
        assigned the wrong substrate specificity. The dehydro-2-deoxygluconokinase (EC 2.7.1.92)
        activity is already encoded by KdgK/b3526, a different member of the same sugar kinase
        superfamily. This is a classic overpropagation mistake (error type 6) - failing to
        distinguish nonisofunctional paralogous subgroups within the superfamily.
      supported_by:
        - reference_id: PMID:40703034
          supporting_text: "b2100 was annotated as a dehydro-2-deoxygluconokinase (EC 2.7.1.92) using DeepECTF...The dehydro-2-deoxygluconokinase (EC 2.7.1.92) activity is encoded by another member of this superfamily KdgK/b3526"
        - reference_id: PMID:40703034
          supporting_text: "Here, DeepECTF predicted correctly the first 3 digits of the EC number but not the last, making an overpropagation mistake"