View original ARBA rule on UniProt
Deprecated/non-existent ARBA rule allegedly applying GO:0160103 (tRNA (guanine(26)-N2/guanine(27)-N2)-dimethyltransferase activity) to TRM1/Trm1 enzymes
Condition-set counts describe the sets recorded in this review, which may omit the full rule.
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.
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.
The dual G26/G27 specificity is not supported by evidence for eukaryotic enzymes and adds unnecessary complexity that contradicts well-established biochemistry
The primary cited evidence (PMID:9801306) explicitly contradicts dual activity, showing only G26 specificity
Rule appears to be deprecated; cannot assess condition overlap
GO:0160103 (dual G26/G27 activity) is too broad; GO:0160104 (G26-specific) is more appropriate for characterized eukaryotic enzymes
If the dual activity term is retained, it should be restricted to specific lineages where G27 activity is documented, not applied broadly to eukaryotes
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