id: ARBA00088072
description: "Deprecated/non-existent ARBA rule allegedly applying GO:0160103 (tRNA (guanine(26)-N2/guanine(27)-N2)-dimethyltransferase activity) to TRM1/Trm1 enzymes"
status: COMPLETE
rule_type: ARBA
rule:
  rule_id: ARBA00088072
  condition_sets: []
  go_annotations:
  - go_id: GO:0160103
    go_label: "tRNA (guanine(26)-N2/guanine(27)-N2)-dimethyltransferase activity"
    aspect: MF
  reviewed_protein_count: 0
  unreviewed_protein_count: 0
  created_date: "UNKNOWN"
  modified_date: "UNKNOWN"
  entries: []

review_summary: |
  This analysis reveals a significant curation problem involving TRM1/Trm1 tRNA 
  dimethyltransferases and conflicting GO term applications. Key findings:

  1. **Rule Status**: ARBA00088072 is not found in current UniProt ARBA database,
     suggesting deprecation or misidentification.

  2. **Literature Evidence**: The primary literature (PMID:9801306) definitively shows
     yeast TRM1 as a G26-specific dimethyltransferase (m22G26), with no evidence for
     G27 modification. The paper's title and abstract explicitly reference G26 specificity.

  3. **GO Term Analysis**:
     - GO:0160103: tRNA (guanine(26)-N2/guanine(27)-N2)-dimethyltransferase activity
       (requires 4 S-adenosyl-L-methionine, dual specificity)
     - GO:0160104: tRNA (guanine(26)-N2)-dimethyltransferase activity  
       (requires 2 S-adenosyl-L-methionine, G26-specific)

  4. **Current Annotation Issues**: S. cerevisiae TRM1 (P15565) is incorrectly annotated
     with both GO:0160103 (from SGD with IDA/IMP evidence citing PMID:9801306) and
     GO:0160104 (from phylogenetic analysis). The GO:0160103 annotation is not supported
     by the cited literature.

  5. **Biological Accuracy**: Extensive research confirms that eukaryotic TRM1/Trm1 
     enzymes across species (yeast, human, fission yeast) are G26-specific. G27 
     methylation activity, if it exists, may be limited to certain bacterial/archaeal
     lineages and is poorly characterized.

  6. **Impact Assessment**: This represents a significant over-annotation problem that
     has propagated incorrect functional information to well-characterized enzymes.

action: DEPRECATE
action_rationale: |
  This ARBA rule appears to be deprecated or non-existent in the current UniProt database.
  More critically, the biological basis for the rule was fundamentally flawed. The rule allegedly
  applied GO:0160103 (dual G26/G27 dimethyltransferase activity) to TRM1/Trm1 enzymes, but
  extensive literature analysis demonstrates that eukaryotic TRM1/Trm1 enzymes are highly 
  specific for G26 dimethylation only. The key evidence cited (PMID:9801306) explicitly 
  describes G26-specific activity with no mention of G27 modification. The dual activity
  term (GO:0160103) represents a significant over-annotation that propagated incorrect 
  functional information.

suggested_modifications:
- "Remove GO:0160103 annotations from TRM1/Trm1 orthologs in eukaryotes"
- "Retain GO:0160104 annotations for eukaryotic TRM1/Trm1 enzymes"
- "Review all proteins currently annotated with GO:0160103 for evidence quality"
- "Consider adding taxonomic restrictions to GO:0160103 if dual activity is documented in specific lineages"
- "Prevent creation of similar rules without stronger evidence base"

parsimony:
  assessment: OVERLY_COMPLEX
  notes: "The dual G26/G27 specificity is not supported by evidence for eukaryotic enzymes and adds unnecessary complexity that contradicts well-established biochemistry"
  supported_by:
  - reference_id: "PMID:9801306"
    supporting_text: "Point and deletion mutations eliminate one or both methyl group transfers catalysed by the yeast TRM1 encoded tRNA (m22G26)dimethyltransferase"

literature_support:
  assessment: CONTRADICTED
  notes: "The primary cited evidence (PMID:9801306) explicitly contradicts dual activity, showing only G26 specificity"
  supported_by:
  - reference_id: "PMID:9801306"
    supporting_text: "Guanosine at position 26 in eukaryotic tRNAs is usually modified to N2 , N2 -dimethylguanosine (m22G26). In Saccharomyces cerevisiae , this reaction is catalysed by the TRM1 encoded tRNA (m22G26)dimethyltransferase"
  - reference_id: "https://github.com/geneontology/go-annotation/issues/5782"
    supporting_text: "The 2 SGD annotations are from the same paper PMID:9801306 This paper doesn't mention G27?"

condition_overlap:
  assessment: NONE
  notes: "Rule appears to be deprecated; cannot assess condition overlap"

go_specificity:
  assessment: TOO_BROAD
  notes: "GO:0160103 (dual G26/G27 activity) is too broad; GO:0160104 (G26-specific) is more appropriate for characterized eukaryotic enzymes"
  supported_by:
  - reference_id: "https://github.com/geneontology/go-annotation/issues/5782"
    supporting_text: "Most characterized Trm1 enzymes are primarily responsible for methylation at G26. In Saccharomyces cerevisiae (budding yeast) and humans, the Trm1 homologs are mostly described as modifying only G26."

taxonomic_scope:
  assessment: TOO_BROAD
  notes: "If the dual activity term is retained, it should be restricted to specific lineages where G27 activity is documented, not applied broadly to eukaryotes"
  supported_by:
  - reference_id: "https://github.com/geneontology/go-annotation/issues/5782"
    supporting_text: "Some bacterial/archaeal Trm1-like enzymes Possibly G26 and G27 Less common, needs more study"

confidence: 0.95