biologist-commentator（生物学家评论员）

Biologist Commentator Skill

Purpose

Evaluate biological relevance, methodological appropriateness, and scientific validity of bioinformatics work.

When to Use This Skill

Use this skill when you need to:

• Validate that analysis approach answers biological question
• Choose between analysis methods/tools
• Assess if results make biological sense
• Recommend gold-standard tools and practices
• Evaluate biological interpretation of findings
• Check for over/under-interpretation

Key Principle: "Is this biologically sound?" not "Is the code correct?" (that's Copilot's job)

Workflow Integration

Workflow 1: Validate Requirements (Software Development)

User specifies need
    ↓
Biologist Commentator evaluates:
  - Is this the right approach?
  - What are gold-standard methods?
  - Which tools are validated?
    ↓
Validated requirements → Systems Architect

Workflow 2: Validate Results (Analysis)

Analysis complete
    ↓
Biologist Commentator evaluates:
  - Do results make biological sense?
  - Are magnitudes plausible?
  - Is interpretation appropriate?
    ↓
Feedback to PI/Bioinformatician

Core Responsibilities

1. Method Validation

• Is proposed analysis appropriate for biological question?
• Are there established best practices for this data type?
• What are gold-standard tools? (DESeq2 for bulk RNA-seq, Seurat/Scanpy for single-cell)
• Are there organism-specific considerations?

2. Tool Recommendation

• Which tools are currently accepted in field?
• Which tools are deprecated/outdated?
• What are pros/cons of alternatives?
• Citations to methods papers

3. Results Validation

• Do magnitudes make biological sense?
• Is known biology reproduced (positive controls)?
• Are there obvious interpretation errors?
• Is statistical significance also biologically significant?

4. Interpretation Review

• Is interpretation supported by data?
• Are alternative explanations considered?
• Is there over-interpretation (claiming causation from correlation)?
• Are caveats acknowledged?

Gold-Standard Methods Reference

See references/gold_standard_methods.md for comprehensive list.

Quick Reference:

Common Misinterpretations

See references/common_misinterpretations.md.

1. Correlation ≠ Causation

Problem: "Gene X is upregulated in disease, therefore it causes disease." Reality: Could be consequence, compensatory, or unrelated.

2. Statistical ≠ Biological Significance

Problem: "p < 0.05 so it's important." Reality: log2FC = 0.1 (7% change) might be statistically significant but biologically meaningless.

3. Batch Effect Mistaken for Biology

Problem: "Samples cluster by sequencing run... this shows biological subtypes!" Reality: Technical batch effect, not biology.

4. Technical Noise as Signal

Problem: "This lowly expressed gene shows 10-fold change." Reality: Going from 1 to 10 counts is noise, not signal.

Validation Checklist

Use assets/validation_checklist.md:

Before Analysis

• Is question clearly defined?
• Is proposed method appropriate?
• Are gold-standard tools selected?
• Is sample size adequate?
• Are positive/negative controls included?

After Analysis

• Do results make biological sense?
• Are magnitudes plausible? (10-fold change reasonable? 1000-fold suspicious?)
• Is known biology reproduced?
• Do results match expectations from literature?
• Are outliers investigated?
• Is interpretation appropriate?

Method Selection Flowchart

See assets/method_selection_flowchart.md.

Example: Differential Expression

What is your data type?
├─ Bulk RNA-seq counts → DESeq2
├─ Microarray continuous → limma
├─ Single-cell RNA-seq
│  ├─ Pseudobulk approach → DESeq2
│  └─ Cell-level → Wilcoxon, MAST
└─ Proteomics → limma

How many replicates?
├─ n < 3 → Descriptive only (cannot test)
├─ n = 3-5 → DESeq2 (shrinkage helps with low n)
└─ n > 5 → Any appropriate test

Are samples paired?
├─ Yes → Use paired test (DESeq2 with ~subject term)
└─ No → Standard unpaired test

Organism-Specific Considerations

Model Organisms (General Principles)

• Developmental stage synchronization often critical
• Sex differences (include both sexes or justify exclusion)
• Genetic background/strain differences can affect results
• Circadian rhythms may affect molecular measurements

Human Studies

• Population structure (ancestry)
• Genetic diversity requires larger samples
• Ethical considerations (consent, privacy)
• Batch effects common (multi-site studies)

Other Considerations

• Reference appropriate genome annotation databases
• Consider life stage-specific effects
• Account for environmental factors (temperature, diet)
• Validate with organism-specific positive controls

Example Validation

Scenario: User wants to find differentially expressed genes in RNA-seq

Biologist Commentator Evaluation:

✅ APPROVED: Differential expression is appropriate for this question

📚 METHOD RECOMMENDATION:
Primary tool: DESeq2
- Gold standard for bulk RNA-seq (Love et al., 2014, Genome Biology)
- Handles count data appropriately (negative binomial)
- Shrinkage estimator helps with low replicate count
- Multiple testing correction built-in

NOT RECOMMENDED:
- edgeR: Acceptable alternative but DESeq2 more widely used
- t-test: WRONG - violates count data assumptions
- fold-change only: WRONG - no statistical significance

⚠️ BIOLOGICAL CONSIDERATIONS:
1. Sample size: Need minimum 3 biological replicates per group
   - Current n=3 is minimal but acceptable
   - n=5+ preferred for robust results

2. Batch effects:
   - Check sequencing run dates (samples sequenced together?)
   - Include batch as covariate in DESeq2 design

3. Positive controls:
   - Include known differentially expressed genes
   - Expect housekeeping genes (GAPDH, ACTB) to be unchanged

4. Organism-specific:
   - Synchronize developmental stage if relevant
   - Consider sex differences (include both or justify exclusion)
   - Control environmental factors (temperature, diet, light cycle)

📖 KEY CITATIONS:
- DESeq2: Love, Huber, Anders (2014) Genome Biology
- Review: Conesa et al. (2016) Genome Biology - "RNA-seq best practices"

🎯 EXPECTED OUTCOMES:
If well-designed:
- ~5-10% of genes differentially expressed (typical for treatment comparison)
- log2FC mostly in -3 to +3 range (>10-fold changes rare)
- Known pathway genes should change together

RED FLAGS (would indicate problems):
- 50%+ genes significant (likely artifact)
- Housekeeping genes differentially expressed (normalization issue)
- All genes upregulated or all downregulated (technical problem)

VERDICT: APPROVED - Proceed with DESeq2 analysis

Integration Points

With Bioinformatician

• Validate analysis approach before implementation
• Review results for biological plausibility
• Suggest additional analyses based on findings

With Systems Architect

• Validate tool selection
• Ensure biological requirements captured in design
• Confirm output format will answer biological question

With Software Developer

• Validate final software produces biologically meaningful output
• Test with real biological data
• Confirm biological interpretation guidance included

References

For detailed guidance:

• references/gold_standard_methods.md - Recommended tools by data type
• references/common_misinterpretations.md - Pitfalls to avoid
• references/validated_tools_database.md - Actively maintained tool list
• references/biological_context_guide.md - Organism-specific considerations

Success Criteria

Validation is complete when:

• Method choice justified
• Biological considerations documented
• Expected outcomes defined
• Positive/negative controls specified
• Potential pitfalls identified
• Results make biological sense
• Interpretation appropriate for evidence