ARBA00026647

View original ARBA rule on UniProt

Type: ARBA
Status: COMPLETE
Action: DEPRECATE
Confidence: 0.10

Description

Mega-rule predicting GO:0010629 negative regulation of gene expression across 131 condition sets covering diverse RNA-binding proteins, chromatin modifiers, nucleases, and regulatory complexes including Dicer, PIWI, SET domain histone methyltransferases, sirtuins, pumilio repeats, and zinc finger proteins

Analysis Summary

Condition-set counts describe the sets recorded in this review, which may omit the full rule.

0
Domain Pairs Analyzed
0
Recorded condition sets
0
Subset Relationships
0
Redundant Annotations

Review Summary

This is an extremely complex mega-rule with 131 condition sets that attempts to capture the broad biological concept of negative gene expression regulation. While the GO term GO:0010629 is biologically valid and many included protein families do participate in gene silencing, the rule suffers from severe over-complexity that undermines its utility and accuracy. The rule conflates distinct molecular mechanisms (RNA interference, chromatin modification, transcriptional repression, mRNA decay) into a single annotation, creating high risk for false positives and loss of biological specificity.

Action Rationale

The rule should be deprecated due to fundamental design flaws: (1) Excessive complexity with 131 condition sets makes it unmaintainable and analysis-prohibitive, (2) Over-broad biological scope conflating mechanistically distinct pathways, (3) Documented false positives such as ESA1 histone acetyltransferase being incorrectly annotated as a negative regulator when it primarily activates gene expression, (4) High false positive risk from promiscuous domains like zinc fingers and helicases that appear in many non-regulatory contexts, (5) Loss of biological specificity by using overly general GO term when more specific child terms exist for different mechanisms, (6) Taxonomic over-annotation applying complex eukaryotic regulatory mechanisms across all domains of life

Rule Definition

Assessments

OVERLY_COMPLEX

Rule exhibits extreme complexity with 131 condition sets, far exceeding the recommended maximum of 12. This level of complexity suggests the rule attempts to capture too broad a biological concept and should be decomposed into multiple focused rules. Many condition sets likely overlap in their protein targets, indicating redundancy.

MODERATE

Individual protein families included in this rule (Dicer, PIWI, SET domains, sirtuins, pumilio repeats) are well-established negative regulators of gene expression with strong literature support. However, the rule conflates mechanistically distinct pathways and incorrectly includes activating enzymes like acetyltransferases. The ESA1 example demonstrates this fundamental flaw: histone acetyltransferases activate rather than repress gene expression.

Supporting Evidence:

  • file:rules/arba/ARBA00026647/ARBA00026647-deep-research-manual.md: ESA1 (Histone Acetyltransferase): This rule incorrectly annotates S. cerevisiae ESA1 (UniProt Q08649) with GO:0010629 negative regulation of gene expression. However, ESA1 is a histone H4 acetyltransferase that primarily ACTIVATES gene expression through chromatin acetylation. This represents a clear false positive where the rule conflates chromatin-modifying enzymes regardless of their activating vs. repressive roles.
SIGNIFICANT

With 131 condition sets, significant overlap is mathematically inevitable. The rule includes structurally related domains (multiple helicase domains, various zinc finger types) and functionally related protein families that likely share protein targets, indicating substantial redundancy.

TOO_BROAD

GO:0010629 negative regulation of gene expression is overly broad for a rule covering mechanistically distinct processes. More specific child terms exist for different regulatory mechanisms: GO:0031047 (gene silencing by RNA), GO:0016458 (gene silencing), GO:0045814 (negative regulation of transcription by chromatin remodeling), GO:0017148 (negative regulation of translation)

TOO_BROAD

Rule applies complex eukaryotic gene silencing mechanisms across inappropriate taxonomic ranges. RNAi machinery (Dicer, PIWI) is primarily eukaryotic, chromatin modifications are largely eukaryotic, yet some condition sets lack taxonomic restrictions while others are inconsistently applied.

References (2)

Raw YAML

View Source YAML
id: ARBA00026647
description: 'Mega-rule predicting GO:0010629 negative regulation of gene expression across 131 condition sets covering diverse RNA-binding proteins, chromatin modifiers, nucleases, and regulatory complexes including Dicer, PIWI, SET domain histone methyltransferases, sirtuins, pumilio repeats, and zinc finger proteins'
status: COMPLETE
rule_type: ARBA
rule:
  rule_id: ARBA00026647
  condition_sets: []
  go_annotations: []  # Rule targets GO:0010629 negative regulation of gene expression
  reviewed_protein_count: 0
  unreviewed_protein_count: 0
  created_date: ''
  modified_date: ''
  entries: []
review_summary: 'This is an extremely complex mega-rule with 131 condition sets that attempts to capture the broad biological concept of negative gene expression regulation. While the GO term GO:0010629 is biologically valid and many included protein families do participate in gene silencing, the rule suffers from severe over-complexity that undermines its utility and accuracy. The rule conflates distinct molecular mechanisms (RNA interference, chromatin modification, transcriptional repression, mRNA decay) into a single annotation, creating high risk for false positives and loss of biological specificity.'
action: DEPRECATE
action_rationale: 'The rule should be deprecated due to fundamental design flaws: (1) Excessive complexity with 131 condition sets makes it unmaintainable and analysis-prohibitive, (2) Over-broad biological scope conflating mechanistically distinct pathways, (3) Documented false positives such as ESA1 histone acetyltransferase being incorrectly annotated as a negative regulator when it primarily activates gene expression, (4) High false positive risk from promiscuous domains like zinc fingers and helicases that appear in many non-regulatory contexts, (5) Loss of biological specificity by using overly general GO term when more specific child terms exist for different mechanisms, (6) Taxonomic over-annotation applying complex eukaryotic regulatory mechanisms across all domains of life'
suggested_modifications:
- 'Replace with multiple smaller, mechanistically focused rules (e.g., separate rules for RNAi, chromatin silencing, transcriptional repression)'
- 'Use more specific GO terms like GO:0031047 (gene silencing by RNA), GO:0045814 (negative regulation by chromatin remodeling), GO:0045892 (negative regulation of transcription)'
- 'Add stricter domain architecture requirements to reduce false positives'
- 'Implement appropriate taxonomic restrictions for lineage-specific mechanisms'
- 'Require multiple domains per condition set to increase specificity'
parsimony:
  assessment: OVERLY_COMPLEX
  notes: 'Rule exhibits extreme complexity with 131 condition sets, far exceeding the recommended maximum of 12. This level of complexity suggests the rule attempts to capture too broad a biological concept and should be decomposed into multiple focused rules. Many condition sets likely overlap in their protein targets, indicating redundancy.'
  supported_by:
  - reference_id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
    supporting_text: 'Rule contains 131 condition sets covering diverse protein families from RNase III domains to zinc fingers to SET domain methyltransferases'
literature_support:
  assessment: MODERATE
  notes: 'Individual protein families included in this rule (Dicer, PIWI, SET domains, sirtuins, pumilio repeats) are well-established negative regulators of gene expression with strong literature support. However, the rule conflates mechanistically distinct pathways and incorrectly includes activating enzymes like acetyltransferases. The ESA1 example demonstrates this fundamental flaw: histone acetyltransferases activate rather than repress gene expression.'
  supported_by:
  - reference_id: file:rules/arba/ARBA00026647/ARBA00026647-deep-research-manual.md
    supporting_text: 'ESA1 (Histone Acetyltransferase): This rule incorrectly annotates S. cerevisiae ESA1 (UniProt Q08649) with GO:0010629 negative regulation of gene expression. However, ESA1 is a histone H4 acetyltransferase that primarily ACTIVATES gene expression through chromatin acetylation. This represents a clear false positive where the rule conflates chromatin-modifying enzymes regardless of their activating vs. repressive roles.'
condition_overlap:
  assessment: SIGNIFICANT
  notes: 'With 131 condition sets, significant overlap is mathematically inevitable. The rule includes structurally related domains (multiple helicase domains, various zinc finger types) and functionally related protein families that likely share protein targets, indicating substantial redundancy.'
  supported_by:
  - reference_id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
    supporting_text: 'Multiple related domains present including various zinc finger types (CCHC, CCCH, 3CxxC), helicase domains, and RNA-binding repeats that commonly co-occur in regulatory proteins'
go_specificity:
  assessment: TOO_BROAD
  notes: 'GO:0010629 negative regulation of gene expression is overly broad for a rule covering mechanistically distinct processes. More specific child terms exist for different regulatory mechanisms: GO:0031047 (gene silencing by RNA), GO:0016458 (gene silencing), GO:0045814 (negative regulation of transcription by chromatin remodeling), GO:0017148 (negative regulation of translation)'
  supported_by:
  - reference_id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
    supporting_text: 'Rule includes diverse mechanisms: RNase III/Dicer (RNAi pathway), SET domains (chromatin silencing), PIWI proteins (piRNA silencing), sirtuins (transcriptional regulation), mRNA decapping enzymes (post-transcriptional control)'
taxonomic_scope:
  assessment: TOO_BROAD
  notes: 'Rule applies complex eukaryotic gene silencing mechanisms across inappropriate taxonomic ranges. RNAi machinery (Dicer, PIWI) is primarily eukaryotic, chromatin modifications are largely eukaryotic, yet some condition sets lack taxonomic restrictions while others are inconsistently applied.'
  supported_by:
  - reference_id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
    supporting_text: 'Taxonomic restrictions are inconsistent: some condition sets restricted to Metazoa, Eukaryota, or specific lineages (Glires, Saccharomyces), while others lack restrictions despite containing eukaryote-specific domains'
confidence: 0.1
references:
- id: file:rules/arba/ARBA00026647/ARBA00026647-deep-research-manual.md
  title: Manual deep research on ARBA00026647 negative regulation mechanisms
  findings:
  - statement: 'GO:0010629 encompasses multiple distinct molecular mechanisms including transcriptional repression, RNA interference, chromatin-mediated silencing, and post-transcriptional control'
  - statement: 'Individual protein families (Dicer, PIWI, SET domains, sirtuins) have strong literature support for gene silencing functions'
  - statement: 'Rule conflates mechanistically distinct pathways that should use more specific GO terms'
  - statement: 'Many included domains (zinc fingers, helicases) are promiscuous and appear in non-regulatory contexts'
  - statement: 'Complex eukaryotic mechanisms like RNAi are inappropriately applied across taxonomic ranges'
- id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
  title: ARBA rule enriched data
  findings:
  - statement: 'Rule contains 131 condition sets targeting GO:0010629 negative regulation of gene expression'
  - statement: 'Includes diverse protein families: RNase III domains, helicases, PIWI domains, SET domains, sirtuins, pumilio repeats, zinc fingers, and RNA-binding proteins'
  - statement: 'Taxonomic restrictions are inconsistently applied across condition sets'
  - statement: 'Rule complexity exceeds analysis capabilities (>12 condition set limit)'
supported_by:
- reference_id: file:rules/arba/ARBA00026647/ARBA00026647-deep-research-manual.md
  supporting_text: 'The rule should be DEPRECATED and replaced with multiple mechanistically focused rules using more specific GO terms: RNAi pathway rule → GO:0031047 gene silencing by RNA, Chromatin silencing rule → GO:0016458 gene silencing or GO:0045814 negative regulation of transcription DNA-templated, Post-transcriptional control rule → GO:0017148 negative regulation of translation'
- reference_id: file:rules/arba/ARBA00026647/ARBA00026647.enriched.json
  supporting_text: 'Analysis of rule structure reveals 131 condition sets with diverse protein domain combinations and inconsistent taxonomic restrictions targeting broad GO term for gene expression regulation'