@wentorai/research-plugins 1.0.0 → 1.2.0

package/skills/domains/biomedical/clinical-dialogue-agents-guide/SKILL.md

@@ -0,0 +1,145 @@
+---
+name: clinical-dialogue-agents-guide
+description: "Papers on AI agents for clinical dialogue and medical QA"
+metadata:
+  openclaw:
+    emoji: "🗣️"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["clinical dialogue", "medical QA", "patient interaction", "clinical agents", "healthcare AI", "diagnosis"]
+    source: "https://github.com/xqz614/Awesome-Agentic-Clinical-Dialogue"
+---
+
+# Agentic Clinical Dialogue Guide
+
+## Overview
+
+A curated collection of papers on AI agents for clinical dialogue — systems that conduct patient interviews, perform differential diagnosis, explain medical information, and support clinical decision-making through conversation. Covers medical QA benchmarks, patient simulation, clinical reasoning chains, and safety considerations unique to healthcare AI.
+
+## Research Landscape
+
+```
+Agentic Clinical Dialogue
+├── Patient-Facing Agents
+│   ├── Symptom checkers
+│   ├── Triage systems
+│   ├── Health information
+│   └── Follow-up management
+├── Clinician-Facing Agents
+│   ├── Diagnostic support
+│   ├── Treatment recommendation
+│   ├── Clinical documentation
+│   └── Literature integration
+├── Clinical Reasoning
+│   ├── Differential diagnosis
+│   ├── History taking
+│   ├── Physical exam interpretation
+│   └── Test ordering
+├── Patient Simulation
+│   ├── Standardized patients (SP)
+│   ├── Medical education
+│   └── Agent evaluation
+└── Safety & Ethics
+    ├── Hallucination in medicine
+    ├── Bias in clinical AI
+    ├── Liability frameworks
+    └── Informed consent
+```
+
+## Key Systems
+
+| System | Focus | Approach |
+|--------|-------|----------|
+| **AMIE** | Diagnostic dialogue | LLM with clinical reasoning |
+| **Med-PaLM** | Medical QA | Finetuned on medical data |
+| **ChatDoctor** | Patient consultation | LLaMA + medical knowledge |
+| **AgentClinic** | Clinical evaluation | Simulated clinical encounters |
+| **ClinicalAgent** | Decision support | Multi-step clinical reasoning |
+
+## Benchmarks
+
+```python
+benchmarks = {
+    "MedQA (USMLE)": {
+        "task": "US Medical Licensing Exam questions",
+        "size": "11,450 questions",
+        "metric": "Accuracy",
+    },
+    "PubMedQA": {
+        "task": "Biomedical yes/no/maybe QA",
+        "size": "1,000 expert-labeled",
+        "metric": "Accuracy",
+    },
+    "AgentClinic": {
+        "task": "Simulated clinical encounters",
+        "size": "Various patient scenarios",
+        "metric": "Diagnostic accuracy + safety",
+    },
+    "MedMCQA": {
+        "task": "Indian medical entrance MCQs",
+        "size": "194k questions",
+        "metric": "Accuracy",
+    },
+    "HealthSearchQA": {
+        "task": "Consumer health search questions",
+        "size": "3,375 questions",
+        "metric": "Expert evaluation",
+    },
+}
+
+for name, info in benchmarks.items():
+    print(f"\n{name}:")
+    print(f"  Task: {info['task']}")
+    print(f"  Size: {info['size']}")
+```
+
+## Safety Considerations
+
+```markdown
+### Critical Safety Issues
+1. **Hallucination** — Fabricated medical facts are dangerous
+2. **Scope limitations** — AI must know when to defer to human
+3. **Emergency recognition** — Must identify urgent situations
+4. **Bias** — Demographic biases in training data
+5. **Liability** — Legal framework for AI medical advice
+6. **Privacy** — Patient data protection (HIPAA compliance)
+
+### Safety Patterns
+- Always recommend consulting healthcare providers
+- Flag emergency symptoms immediately
+- Disclose AI nature to patients
+- Log all interactions for audit
+- Implement uncertainty quantification
+```
+
+## Reading Roadmap
+
+```markdown
+### Foundations
+1. AMIE: "Towards Conversational Diagnostic AI" (Google, 2024)
+2. Med-PaLM 2: "Expert-level medical QA" (Google, 2023)
+3. "Evaluating LLMs in Clinical Dialogue" (Survey, 2024)
+
+### Clinical Reasoning
+4. "Chain-of-Diagnosis" (Clinical CoT, 2024)
+5. "AgentClinic: Evaluating Clinical Agents" (2024)
+6. "Simulated Patient Encounters with LLMs" (2024)
+
+### Safety
+7. "Hallucination in Medical AI" (Survey, 2024)
+8. "Red Teaming Medical LLMs" (2024)
+```
+
+## Use Cases
+
+1. **Research survey**: Map clinical dialogue AI landscape
+2. **Benchmark tracking**: Compare medical AI performance
+3. **System design**: Learn from clinical agent architectures
+4. **Safety analysis**: Understand risks and mitigations
+5. **Medical education**: Patient simulation for training
+
+## References
+
+- [Awesome-Agentic-Clinical-Dialogue](https://github.com/xqz614/Awesome-Agentic-Clinical-Dialogue)
+- [AMIE Paper](https://arxiv.org/abs/2401.05654)
+- [AgentClinic](https://arxiv.org/abs/2405.07960)

package/skills/domains/biomedical/ena-sequence-api/SKILL.md

@@ -0,0 +1,175 @@
+---
+name: ena-sequence-api
+description: "Access nucleotide sequence data from the European Nucleotide Archive"
+metadata:
+  openclaw:
+    emoji: "🔬"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["ENA", "nucleotide sequences", "genomics", "EMBL-EBI", "sequencing data", "NGS"]
+    source: "https://www.ebi.ac.uk/ena/"
+---
+
+# European Nucleotide Archive (ENA) API
+
+## Overview
+
+The European Nucleotide Archive (ENA) at EMBL-EBI is one of the three global nucleotide sequence databases (with NCBI GenBank and DDBJ). It provides access to raw sequencing reads, assembled sequences, and functional annotations from all organisms. The API supports accession lookup, text search, and bulk data retrieval. Free, no authentication required.
+
+## API Endpoints
+
+### Portal API (Search)
+
+```bash
+# Search for studies
+curl "https://www.ebi.ac.uk/ena/portal/api/search?query=CRISPR+cas9&result=study&limit=20&format=json"
+
+# Search for samples
+curl "https://www.ebi.ac.uk/ena/portal/api/search?query=human+gut+microbiome&result=sample&limit=20&format=json"
+
+# Search for runs (sequencing data)
+curl "https://www.ebi.ac.uk/ena/portal/api/search?query=RNA-seq+cancer&result=read_run&limit=20&format=json"
+```
+
+### Browser API (Accession Lookup)
+
+```bash
+# Get record by accession
+curl "https://www.ebi.ac.uk/ena/browser/api/xml/PRJEB12345"
+
+# Get in JSON format
+curl "https://www.ebi.ac.uk/ena/browser/api/summary/PRJEB12345"
+
+# Get sequence in FASTA
+curl "https://www.ebi.ac.uk/ena/browser/api/fasta/AF123456"
+
+# Get in EMBL flat file format
+curl "https://www.ebi.ac.uk/ena/browser/api/embl/AF123456"
+```
+
+### Taxonomy Search
+
+```bash
+# Search by organism
+curl "https://www.ebi.ac.uk/ena/portal/api/search?query=tax_tree(9606)&result=study&limit=20&format=json"
+
+# Get taxonomy details
+curl "https://www.ebi.ac.uk/ena/taxonomy/rest/tax-id/9606"
+```
+
+### Result Types
+
+| Type | Description | Example accession |
+|------|-------------|-------------------|
+| `study` | Research project | PRJEB12345 |
+| `sample` | Biological sample | SAMEA12345 |
+| `experiment` | Library/protocol | ERX12345 |
+| `read_run` | Sequencing run | ERR12345 |
+| `analysis` | Computed analysis | ERZ12345 |
+| `sequence` | Assembled sequence | AF123456 |
+| `wgs_set` | Whole genome shotgun | AABR00000000 |
+
+### Query Parameters
+
+| Parameter | Description | Example |
+|-----------|-------------|---------|
+| `query` | Search text or taxonomy | `query=SARS-CoV-2` |
+| `result` | Result type | `result=study` |
+| `limit` | Max results (default 100K) | `limit=50` |
+| `offset` | Pagination offset | `offset=100` |
+| `format` | Response format | `json`, `tsv`, `xml` |
+| `fields` | Specific fields | `fields=accession,description` |
+
+## Python Usage
+
+```python
+import requests
+
+PORTAL_URL = "https://www.ebi.ac.uk/ena/portal/api"
+BROWSER_URL = "https://www.ebi.ac.uk/ena/browser/api"
+
+
+def search_studies(query: str, limit: int = 20) -> list:
+    """Search ENA for research studies."""
+    params = {
+        "query": query,
+        "result": "study",
+        "limit": limit,
+        "format": "json",
+        "fields": "study_accession,study_title,study_description,"
+                  "tax_id,scientific_name,center_name",
+    }
+    resp = requests.get(f"{PORTAL_URL}/search", params=params)
+    resp.raise_for_status()
+    return resp.json()
+
+
+def search_runs(query: str, limit: int = 20) -> list:
+    """Search for sequencing runs."""
+    params = {
+        "query": query,
+        "result": "read_run",
+        "limit": limit,
+        "format": "json",
+        "fields": "run_accession,experiment_title,instrument_platform,"
+                  "library_strategy,read_count,base_count",
+    }
+    resp = requests.get(f"{PORTAL_URL}/search", params=params)
+    resp.raise_for_status()
+    return resp.json()
+
+
+def get_fasta(accession: str) -> str:
+    """Retrieve sequence in FASTA format."""
+    resp = requests.get(f"{BROWSER_URL}/fasta/{accession}")
+    resp.raise_for_status()
+    return resp.text
+
+
+def get_study_runs(study_accession: str) -> list:
+    """Get all sequencing runs for a study."""
+    params = {
+        "query": f'study_accession="{study_accession}"',
+        "result": "read_run",
+        "format": "json",
+        "fields": "run_accession,fastq_ftp,read_count,base_count",
+        "limit": 1000,
+    }
+    resp = requests.get(f"{PORTAL_URL}/search", params=params)
+    resp.raise_for_status()
+    return resp.json()
+
+
+# Example: find COVID-19 sequencing studies
+studies = search_studies("SARS-CoV-2 whole genome", limit=5)
+for s in studies:
+    print(f"{s['study_accession']}: {s['study_title']}")
+    print(f"  Organism: {s.get('scientific_name')}")
+
+# Example: find RNA-seq runs
+runs = search_runs("RNA-seq breast cancer", limit=5)
+for r in runs:
+    reads = int(r.get("read_count", 0))
+    print(f"{r['run_accession']}: {r.get('experiment_title', '')}")
+    print(f"  Platform: {r.get('instrument_platform')} | "
+          f"Reads: {reads:,}")
+```
+
+## Data Access
+
+```bash
+# Download FASTQ files (from run metadata)
+# The fastq_ftp field provides FTP paths:
+wget ftp://ftp.sra.ebi.ac.uk/vol1/fastq/ERR123/ERR123456/ERR123456_1.fastq.gz
+
+# Bulk download via Aspera (faster)
+ascp -QT -l 300m -P33001 \
+  era-fasp@fasp.sra.ebi.ac.uk:/vol1/fastq/ERR123/ERR123456/ ./
+```
+
+## References
+
+- [ENA](https://www.ebi.ac.uk/ena/)
+- [ENA Portal API](https://www.ebi.ac.uk/ena/portal/api/)
+- [ENA Browser API](https://www.ebi.ac.uk/ena/browser/api/)
+- Harrison, P.W. et al. (2021). "The European Nucleotide Archive in 2020." *NAR* 49(D1).

package/skills/domains/biomedical/genomas-guide/SKILL.md

@@ -0,0 +1,126 @@
+---
+name: genomas-guide
+description: "Automate gene expression analysis with the GenoMAS multi-agent system"
+metadata:
+  openclaw:
+    emoji: "🧬"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["GenoMAS", "gene expression", "multi-agent", "bioinformatics", "RNA-seq", "genomics automation"]
+    source: "https://github.com/futianfan/GenoMAS"
+---
+
+# GenoMAS Guide
+
+## Overview
+
+GenoMAS (Genomics Multi-Agent System) is a minimalist multi-agent framework for automating scientific analysis workflows, particularly gene expression analysis. It orchestrates specialized agents for data retrieval, preprocessing, differential expression analysis, pathway enrichment, and visualization — turning a natural language research question into a complete bioinformatics pipeline.
+
+## Installation
+
+```bash
+pip install genomas
+# Or from source
+git clone https://github.com/futianfan/GenoMAS.git
+cd GenoMAS && pip install -e .
+```
+
+## Core Workflow
+
+### Natural Language to Pipeline
+
+```python
+from genomas import GenoMAS
+
+geno = GenoMAS(llm_provider="anthropic")
+
+# Describe analysis in natural language
+result = geno.analyze(
+    "Compare gene expression between tumor and normal tissue "
+    "in the TCGA breast cancer dataset. Identify differentially "
+    "expressed genes and run pathway enrichment analysis."
+)
+
+# GenoMAS automatically:
+# 1. Retrieves TCGA-BRCA data via GDC API
+# 2. Normalizes and filters expression data
+# 3. Runs DESeq2-style differential expression
+# 4. Performs GO and KEGG pathway enrichment
+# 5. Generates volcano plots and heatmaps
+```
+
+### Agent Roles
+
+| Agent | Responsibility |
+|-------|---------------|
+| **Data Agent** | Retrieves datasets from GEO, TCGA, ArrayExpress |
+| **Preprocessing Agent** | Quality control, normalization, filtering |
+| **Analysis Agent** | Differential expression, clustering, PCA |
+| **Enrichment Agent** | GO, KEGG, MSigDB pathway analysis |
+| **Visualization Agent** | Plots, heatmaps, volcano plots |
+| **Report Agent** | Generates methods section and results summary |
+
+### Step-by-Step Usage
+
+```python
+from genomas import DataAgent, AnalysisAgent, EnrichmentAgent
+
+# Step 1: Retrieve data
+data_agent = DataAgent()
+dataset = data_agent.fetch("GSE12345", platform="RNA-seq")
+
+# Step 2: Differential expression
+analysis = AnalysisAgent()
+de_results = analysis.differential_expression(
+    dataset,
+    group_col="condition",
+    case="tumor",
+    control="normal",
+    method="deseq2",
+)
+
+# Step 3: Filter significant genes
+sig_genes = de_results[
+    (de_results["padj"] < 0.05) &
+    (abs(de_results["log2FoldChange"]) > 1)
+]
+print(f"Found {len(sig_genes)} differentially expressed genes")
+
+# Step 4: Pathway enrichment
+enrichment = EnrichmentAgent()
+pathways = enrichment.run(
+    gene_list=sig_genes["gene_symbol"].tolist(),
+    databases=["GO_BP", "KEGG", "Reactome"],
+)
+
+# Step 5: Visualize
+from genomas.viz import volcano_plot, pathway_barplot
+volcano_plot(de_results, output="volcano.png")
+pathway_barplot(pathways, top_n=20, output="pathways.png")
+```
+
+## Supported Analyses
+
+| Analysis | Method |
+|----------|--------|
+| Differential expression | DESeq2, edgeR, limma-voom |
+| Clustering | Hierarchical, k-means, UMAP |
+| PCA | Principal component analysis |
+| GO enrichment | Gene Ontology term enrichment |
+| KEGG pathway | KEGG pathway mapping |
+| GSEA | Gene Set Enrichment Analysis |
+| Survival analysis | Kaplan-Meier, Cox regression |
+
+## Data Sources
+
+| Source | Data type |
+|--------|-----------|
+| GEO (NCBI) | Microarray, RNA-seq |
+| TCGA | Cancer genomics |
+| GTEx | Normal tissue expression |
+| ArrayExpress | European expression data |
+
+## References
+
+- [GenoMAS GitHub](https://github.com/futianfan/GenoMAS)
+- Love, M.I. et al. (2014). "Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2." *Genome Biology* 15(12).

package/skills/domains/biomedical/genotex-benchmark-guide/SKILL.md

@@ -0,0 +1,125 @@
+---
+name: genotex-benchmark-guide
+description: "Benchmark for LLM agents on gene expression data analysis"
+metadata:
+  openclaw:
+    emoji: "🧫"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["GenoTEX", "gene expression", "benchmark", "LLM agent", "bioinformatics", "GEO"]
+    source: "https://github.com/Liu-Hy/GenoTEX"
+---
+
+# GenoTEX Benchmark Guide
+
+## Overview
+
+GenoTEX is a benchmark for evaluating LLM-based agents on gene expression data analysis tasks. It provides curated datasets from GEO (Gene Expression Omnibus) with ground-truth analysis pipelines, testing agents on data preprocessing, differential expression, enrichment analysis, and biological interpretation. Published at MLCB 2025 as an oral presentation.
+
+## Benchmark Structure
+
+```
+GenoTEX Benchmark
+├── Data Collection
+│   └── Curated GEO datasets with ground truth
+├── Task Categories
+│   ├── Data preprocessing (QC, normalization)
+│   ├── Differential expression analysis
+│   ├── Gene set enrichment analysis
+│   ├── Clustering and classification
+│   └── Biological interpretation
+├── Evaluation
+│   ├── Code correctness (executes without error)
+│   ├── Statistical validity (appropriate tests)
+│   ├── Result accuracy (vs ground truth)
+│   └── Interpretation quality (biological insight)
+└── Baselines
+    ├── GPT-4 agent
+    ├── Claude agent
+    └── Domain-specific fine-tuned models
+```
+
+## Usage
+
+```python
+from genotex import GenoTEXBenchmark
+
+bench = GenoTEXBenchmark()
+
+# List available tasks
+tasks = bench.list_tasks()
+for task in tasks[:5]:
+    print(f"Task: {task.id}")
+    print(f"  Dataset: {task.geo_accession}")
+    print(f"  Category: {task.category}")
+    print(f"  Difficulty: {task.difficulty}")
+
+# Get a specific task
+task = bench.get_task("GSE12345_DEG")
+print(f"Description: {task.description}")
+print(f"Input files: {task.input_files}")
+print(f"Expected output: {task.expected_output_type}")
+```
+
+## Running Evaluations
+
+```python
+# Evaluate an agent on GenoTEX
+from genotex import evaluate_agent
+
+results = evaluate_agent(
+    agent_fn=my_agent_function,
+    tasks="all",            # or specific task IDs
+    timeout_per_task=300,   # seconds
+)
+
+print(f"Tasks completed: {results.completed}/{results.total}")
+print(f"Code correctness: {results.code_correct_rate:.1%}")
+print(f"Statistical validity: {results.stats_valid_rate:.1%}")
+print(f"Result accuracy: {results.accuracy:.3f}")
+```
+
+## Task Examples
+
+```python
+# Example: Differential Expression Analysis
+task = {
+    "id": "GSE12345_DEG",
+    "description": "Identify differentially expressed genes "
+                   "between treatment and control groups in "
+                   "this RNA-seq dataset.",
+    "input": "GSE12345_counts.csv",  # Raw count matrix
+    "metadata": "GSE12345_metadata.csv",  # Sample info
+    "expected": {
+        "method": "DESeq2 or limma-voom",
+        "output": "DEG table with log2FC, p-value, adj.p",
+        "ground_truth": "GSE12345_deg_truth.csv",
+    },
+}
+
+# Example: Gene Set Enrichment
+task = {
+    "id": "GSE12345_GSEA",
+    "description": "Perform gene set enrichment analysis on "
+                   "the DEGs and identify enriched pathways.",
+    "input": "GSE12345_deg_results.csv",
+    "expected": {
+        "method": "fgsea, clusterProfiler, or enrichR",
+        "output": "Enriched pathways with NES and FDR",
+    },
+}
+```
+
+## Use Cases
+
+1. **Agent evaluation**: Test bioinformatics agents on real tasks
+2. **Method comparison**: Compare LLM agents on genomics
+3. **Benchmark development**: Extend with new GEO datasets
+4. **Teaching**: Standard tasks for bioinformatics education
+5. **Tool development**: Test new analysis pipelines
+
+## References
+
+- [GenoTEX GitHub](https://github.com/Liu-Hy/GenoTEX)
+- [GEO Database](https://www.ncbi.nlm.nih.gov/geo/)
+- [MLCB 2025](https://mlcb.github.io/)