{"skill":{"slug":"arrive-guideline-architect-2","displayName":"Arrive Guideline Architect","summary":"Generate ARRIVE 2.0 compliant animal research protocols with structured experimental design, sample size calculations, and reporting checklists. Ensures tran...","description":"---\nname: arrive-guideline-architect\ndescription: Generate ARRIVE 2.0 compliant animal research protocols with structured \n  experimental design, sample size calculations, and reporting checklists. Ensures \n  transparency, reproducibility, and ethical compliance in in vivo studies.\nallowed-tools: [Read, Write, Bash, Edit]\nlicense: MIT\nmetadata:\n    skill-author: AIPOCH\n---\n\n# ARRIVE Guideline Architect\n\n## Overview\n\nAI-powered protocol design tool that creates publication-ready animal research protocols compliant with ARRIVE 2.0 guidelines (Animal Research: Reporting of In Vivo Experiments). Generates structured documentation for ethical review, transparent reporting, and reproducible science.\n\n**Key Capabilities:**\n- **Protocol Generation**: Complete ARRIVE 2.0 compliant study protocols\n- **Sample Size Calculator**: Statistical power analysis with justification\n- **Compliance Checker**: Validate existing protocols against ARRIVE standards\n- **Randomization Schemes**: Generate and document allocation strategies\n- **Ethics Support**: IACUC protocol templates and animal welfare documentation\n- **Reporting Templates**: Manuscript preparation with required elements\n\n## When to Use\n\n**✅ Use this skill when:**\n- Designing new animal studies requiring ethical approval\n- Preparing IACUC (Institutional Animal Care and Use Committee) applications\n- Writing manuscripts for journals requiring ARRIVE compliance (PLOS, Nature, etc.)\n- Validating existing protocols for transparency and completeness\n- Training researchers on animal research best practices\n- Planning multi-site studies requiring standardized protocols\n- Reviewing protocols for grant applications\n\n**❌ Do NOT use when:**\n- Human clinical trials → Use `clinical-protocol-designer`\n- In vitro studies (cell culture only) → No ARRIVE requirements apply\n- Field studies on wild animals → Use specialized wildlife research guidelines\n- Veterinary clinical cases → Use veterinary case report standards\n- Systematic reviews/meta-analyses → Use PRISMA guidelines\n\n**Integration:**\n- **Upstream**: `sample-size-power-calculator` (statistical design)\n- **Downstream**: `iacuc-protocol-drafter` (ethics submission), `manuscript-prep-assistant` (publication)\n\n## Core Capabilities\n\n### 1. ARRIVE 2.0 Protocol Builder\n\nGenerate complete protocols covering all Essential 10 items:\n\n```python\nfrom scripts.arrive_builder import ARRIVEBuilder\n\nbuilder = ARRIVEBuilder()\n\n# Generate full protocol\nprotocol = builder.generate_protocol(\n    title=\"Efficacy of Compound X in Type 2 Diabetes Mouse Model\",\n    species=\"Mus musculus\",\n    strain=\"db/db\",\n    groups=[\n        {\"name\": \"Control\", \"n\": 15, \"treatment\": \"Vehicle\"},\n        {\"name\": \"Low Dose\", \"n\": 15, \"treatment\": \"10 mg/kg\"},\n        {\"name\": \"High Dose\", \"n\": 15, \"treatment\": \"50 mg/kg\"}\n    ],\n    primary_endpoint=\"Fasting blood glucose reduction\",\n    duration_days=28\n)\n\nprotocol.save(\"protocol.md\")\n```\n\n**Generates:**\n1. **Study Design**: Experimental groups, timelines, endpoints\n2. **Sample Size**: Power calculations with justification\n3. **Inclusion/Exclusion**: Animal selection criteria\n4. **Randomization**: Allocation method (software/hardware)\n5. **Blinding**: Who, when, how blinding implemented\n6. **Outcome Measures**: Primary, secondary, exploratory endpoints\n7. **Statistical Methods**: Analysis plan, software, significance level\n8. **Experimental Animals**: Species, strain, sex, age, weight, source\n9. **Experimental Procedures**: Detailed methods with timing\n10. **Results Reporting**: Data presentation templates\n\n### 2. Sample Size Calculator\n\nStatistical power analysis with ARRIVE-compliant justification:\n\n```python\nfrom scripts.sample_size import SampleSizeCalculator\n\ncalc = SampleSizeCalculator()\n\n# Calculate with effect size\nresult = calc.calculate(\n    test_type=\"two_sample_t_test\",\n    effect_size=0.8,  # Cohen's d\n    alpha=0.05,\n    power=0.80,\n    expected_dropout=0.10  # 10% attrition\n)\n\n# Output: n=26 per group (total 78, accounting for 10% dropout)\n```\n\n**Features:**\n- **Effect Size Selection**: Cohen's d, odds ratio, hazard ratio\n- **Multiple Comparisons**: Bonferroni, FDR corrections\n- **Dropout Adjustment**: Account for expected attrition\n- **Justification Text**: Auto-generate sample size rationale\n- **Power Curves**: Generate power calculations for various sample sizes\n\n### 3. Compliance Validator\n\nCheck existing protocols against ARRIVE 2.0:\n\n```bash\npython scripts/validate.py --input my_protocol.md --format markdown\n```\n\n**Output:**\n```\n✅ Essential 10: 10/10 complete\n⚠️  Recommended Set: 8/15 complete\n   Missing: Data sharing statement, Conflict of interest\n\nDetailed Report:\n- Item 1 (Study Design): Complete\n- Item 2 (Sample Size): Complete  \n- Item 3 (Inclusion Criteria): Missing - add exclusion criteria\n- ...\n```\n\n**Validation Levels:**\n- **Essential 10**: Required for all publications\n- **Recommended Set**: Required by top-tier journals\n- **Journal-Specific**: Custom checks for specific publishers\n\n### 4. Randomization & Blinding Generator\n\nCreate allocation schemes with documentation:\n\n```python\nfrom scripts.randomization import RandomizationGenerator\n\ngen = RandomizationGenerator()\n\n# Generate allocation\nallocation = gen.generate(\n    n_animals=45,\n    n_groups=3,\n    method=\"block_randomization\",  # or \"simple\", \"stratified\"\n    block_size=6,\n    seed=42  # For reproducibility\n)\n\n# Output allocation table\nallocation.save(\"allocation_table.csv\")\nallocation.generate_blinding_key(\"blinding_key.xlsx\")\n```\n\n**Methods Supported:**\n- Simple randomization\n- Block randomization (fixed/random block sizes)\n- Stratified randomization (by sex, age, baseline)\n- Covariate-adaptive minimization\n\n## Common Patterns\n\n### Pattern 1: Drug Efficacy Study\n\n**Template for therapeutic intervention studies:**\n\n```json\n{\n  \"study_type\": \"efficacy\",\n  \"species\": \"Mus musculus\",\n  \"model\": \"Disease model (e.g., db/db diabetic mice)\",\n  \"intervention\": \"Test compound\",\n  \"groups\": [\n    \"Sham control\",\n    \"Disease control (vehicle)\",\n    \"Positive control (reference drug)\",\n    \"Test compound (low dose)\",\n    \"Test compound (high dose)\"\n  ],\n  \"primary_endpoint\": \"Disease biomarker\",\n  \"secondary_endpoints\": [\"Safety markers\", \"Histopathology\"],\n  \"sampling_timepoints\": [\"Baseline\", \"Week 2\", \"Week 4\"]\n}\n```\n\n**Key Considerations:**\n- Include positive control for assay validation\n- Multiple doses to establish dose-response\n- Power calculation based on expected effect size\n- Sample size accounts for disease variability\n\n### Pattern 2: Toxicology Study\n\n**Template for safety assessment:**\n\n```json\n{\n  \"study_type\": \"toxicology\",\n  \"species\": \"Rat\",\n  \"duration\": \"28-day repeat dose\",\n  \"dose_levels\": [\"Vehicle\", \"Low\", \"Mid\", \"High\", \"Limit\"],\n  \"endpoints\": [\n    \"Clinical observations (daily)\",\n    \"Body weight (twice weekly)\",\n    \"Food consumption\",\n    \"Clinical pathology (hematology, chemistry)\",\n    \"Necropsy and organ weights\",\n    \"Histopathology\"\n  ],\n  \"recovery_groups\": true  # 14-day recovery period\n}\n```\n\n**Key Considerations:**\n- Dose selection based on MTD (maximum tolerated dose)\n- Recovery groups for reversibility assessment\n- Comprehensive clinical pathology panels\n- Histopathology on all high-dose and control animals\n\n### Pattern 3: Behavioral Study\n\n**Template for neuroscience/behavioral research:**\n\n```json\n{\n  \"study_type\": \"behavioral\",\n  \"species\": \"C57BL/6 mice\",\n  \"tests\": [\n    \"Open field (anxiety/locomotion)\",\n    \"Elevated plus maze (anxiety)\",\n    \"Novel object recognition (memory)\",\n    \"Fear conditioning (learning)\"\n  ],\n  \"controls\": [\n    \"Positive pharmacological control\",\n    \"Negative control (vehicle)\"\n  ],\n  \"blinding\": \"Video analysis performed blinded\",\n  \"randomization\": \"Latin square design for test order\"\n}\n```\n\n**Key Considerations:**\n- Counterbalance test order (learning effects)\n- Blind video analysis to prevent bias\n- Standardized testing environment (lighting, noise)\n- Experimenter training and reliability testing\n\n### Pattern 4: Surgical Model Study\n\n**Template for procedure-based research:**\n\n```json\n{\n  \"study_type\": \"surgical\",\n  \"procedure\": \"Myocardial infarction (LAD ligation)\",\n  \"species\": \"Sprague-Dawley rats\",\n  \"sham_control\": true,\n  \"perioperative_care\": {\n    \"analgesia\": \"Buprenorphine SR\",\n    \"antibiotics\": \"Enrofloxacin\",\n    \"monitoring\": \"Temperature, respiration, pain scoring\"\n  },\n  \"outcome_measures\": [\n    \"Survival rate\",\n    \"Echocardiography\",\n    \"Histological infarct size\"\n  ],\n  \"humane_endpoints\": [\"Severe distress\", \"Inability to ambulate\"]\n}\n```\n\n**Key Considerations:**\n- Detailed surgical protocol with timing\n- Comprehensive perioperative care\n- Clear humane endpoints (refinement)\n- Sham surgery controls for procedure effects\n- Pain management per IACUC guidelines\n\n## Complete Workflow Example\n\n**From study concept to IACUC submission:**\n\n```bash\n# Step 1: Create study brief\ncat > study_brief.json << EOF\n{\n  \"title\": \"Novel Compound X in Diabetic Nephropathy\",\n  \"species\": \"Mouse\",\n  \"strain\": \"db/db\",\n  \"groups\": 4,\n  \"primary_endpoint\": \"Albuminuria reduction\",\n  \"duration_weeks\": 12\n}\nEOF\n\n# Step 2: Generate protocol\npython scripts/main.py \\\n  --input study_brief.json \\\n  --output protocol.md \\\n  --include-checklist\n\n# Step 3: Calculate sample size\npython scripts/sample_size.py \\\n  --test t_test \\\n  --effect-size 0.8 \\\n  --alpha 0.05 \\\n  --power 0.80 \\\n  --dropout 0.10\n\n# Step 4: Generate randomization\npython scripts/randomize.py \\\n  --n-total 64 \\\n  --n-groups 4 \\\n  --method block \\\n  --output allocation.csv\n\n# Step 5: Validate ARRIVE compliance\npython scripts/validate.py \\\n  --input protocol.md \\\n  --format pdf \\\n  --output compliance_report.pdf\n```\n\n**Output Files:**\n```\noutput/\n├── protocol.md                    # Complete ARRIVE protocol\n├── sample_size_justification.txt  # Statistical rationale\n├── allocation.csv                 # Randomization table\n├── blinding_key.xlsx             # Blinding documentation\n├── compliance_report.pdf         # ARRIVE checklist\n└├── iacuc_supplemental.pdf       # Ethics committee materials\n```\n\n## Quality Checklist\n\n**Pre-Study:**\n- [ ] **CRITICAL**: IACUC approval obtained before starting\n- [ ] Sample size adequately powered (≥80%)\n- [ ] Randomization method documented and reproducible\n- [ ] Blinding plan clear for all assessors\n- [ ] Humane endpoints defined with clear criteria\n- [ ] Inclusion/exclusion criteria prespecified\n\n**During Study:**\n- [ ] Randomization followed without deviations\n- [ ] Blinding maintained (unblinding only for safety)\n- [ ] All animals accounted for (CONSORT-style flow diagram)\n- [ ] Adverse events documented and reported to IACUC\n- [ ] Sample collection at predetermined timepoints\n\n**Reporting:**\n- [ ] All Essential 10 items addressed in manuscript\n- [ ] CONSORT-style flow diagram for animal studies\n- [ ] Raw data available (or sharing statement)\n- [ ] Conflict of interest disclosed\n- [ ] Funding sources acknowledged\n\n## Common Pitfalls\n\n**Design Issues:**\n- ❌ **Inadequate controls** → Cannot distinguish treatment from confounding effects\n  - ✅ Always include appropriate controls (vehicle, positive, sham)\n  \n- ❌ **Convenience sampling** → Selection bias\n  - ✅ Random allocation to treatment groups\n\n- ❌ **Unblinded assessment** → Observer bias\n  - ✅ Blinded outcome assessment whenever possible\n\n**Sample Size Issues:**\n- ❌ **No power calculation** → Underpowered study, false negatives\n  - ✅ Calculate sample size a priori with justification\n\n- ❌ **Ignoring dropout** → Final sample too small\n  - ✅ Account for expected attrition (typically 10-20%)\n\n**Reporting Issues:**\n- ❌ **Selective outcome reporting** → Publication bias\n  - ✅ Pre-register primary and secondary endpoints\n\n- ❌ **Missing animal numbers** → Transparency concerns\n  - ✅ Report n for every analysis\n\n## References\n\nAvailable in `references/` directory:\n\n- `arrive_2.0_guidelines.md` - Official ARRIVE 2.0 checklist and explanations\n- `sample_size_guidelines.md` - Statistical methods for animal studies\n- `species_specific_requirements.md` - Mouse, rat, zebrafish considerations\n- `journal_compliance.md` - Requirements by publisher (Nature, Science, Cell)\n- `statistical_methods.md` - Analysis approaches for common designs\n- `iacuc_templates.md` - Ethics committee application templates\n- `example_protocols.md` - Published compliant protocols as examples\n\n## Scripts\n\nLocated in `scripts/` directory:\n\n- `main.py` - Protocol generation CLI\n- `arrive_builder.py` - Core protocol builder\n- `sample_size.py` - Power analysis calculator\n- `randomization.py` - Allocation scheme generator\n- `validate.py` - ARRIVE compliance checker\n- `checklist_generator.py` - Interactive checklist tool\n- `export.py` - Multi-format output (PDF, Word, Markdown)\n\n## Limitations\n\n- **Template-Based**: Generates standard protocols; highly specialized studies may need customization\n- **No Statistical Analysis**: Calculates sample size but does not perform analysis\n- **No Real-Time Monitoring**: Protocol generation only; does not track actual experiments\n- **Species Coverage**: Optimized for mice and rats; other species may need adaptation\n- **Regulatory Variation**: IACUC requirements vary by institution; may need local customization\n\n---\n\n**🐾 Remember: The 3Rs (Replacement, Reduction, Refinement) are ethical imperatives. This tool supports Reduction (optimal sample sizes) and Refinement (better experimental design), but consider Replacement alternatives (in vitro, in silico) whenever possible.**\n\n## Parameters\n\n| Parameter | Type | Default | Description |\n|-----------|------|---------|-------------|\n| `--interactive` | flag | - | **Interactive mode**: Run wizard with guided prompts (uses `input()` for user interaction). Recommended for first-time users or complex study designs. |\n| `--input` | str | Required | Input JSON file path (batch/automation mode) |\n| `--output` | str | \"protocol.md\" | Output file path |\n| `--validate` | str | Required | Validate existing protocol file |\n| `--checklist` | str | Required | Generate ARRIVE 2.0 checklist |\n| `--format` | str | \"markdown\" | Output format: markdown, pdf, or docx |\n\n**Usage Modes:**\n- **Automation Mode (Recommended for CI/CD)**: Use `--input` with JSON configuration file\n- **Interactive Mode**: Use `--interactive` for guided setup via prompts\n\n**Example - Automation Mode:**\n```bash\n# Create JSON config\ncat > study_config.json << 'EOF'\n{\n  \"title\": \"Diabetes Drug Study\",\n  \"species\": \"Mus musculus\",\n  \"strain\": \"db/db\",\n  \"groups\": 4,\n  \"animals_per_group\": 15\n}\nEOF\n\n# Generate protocol\npython scripts/main.py --input study_config.json --output protocol.md\n```\n\n**Example - Interactive Mode:**\n```bash\n# Launch interactive wizard\npython scripts/main.py --interactive\n```\n","tags":{"latest":"0.1.1"},"stats":{"comments":0,"downloads":539,"installsAllTime":20,"installsCurrent":0,"stars":0,"versions":2},"createdAt":1773395577785,"updatedAt":1778491880948},"latestVersion":{"version":"0.1.1","createdAt":1773395638910,"changelog":"- Initial sample data and documentation added: includes ARRIVE checklist, example protocol, and study brief files.\n- Introduced primary script (`scripts/main.py`) to support protocol generation or workflow tasks.\n- No changes to the core skill logic; this update focuses on supplementary examples and startup resources.","license":"MIT-0"},"metadata":null,"owner":{"handle":"theresayao0614-sudo","userId":"s17ea40c1y75h06x45mjjxwg3583qhxm","displayName":"Theresa","image":"https://avatars.githubusercontent.com/u/259492064?v=4"},"moderation":null}