@wentorai/research-plugins 1.0.0

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

@@ -0,0 +1,200 @@
+---
+name: epidemiology-guide
+description: "Epidemiological study designs, measures of association, and public health ana..."
+metadata:
+  openclaw:
+    emoji: "microscope"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["epidemiology", "public health", "evidence-based medicine", "clinical medicine", "disease surveillance"]
+    source: "wentor"
+---
+
+# Epidemiology Guide
+
+A skill for designing and analyzing epidemiological studies. Covers study design selection, measures of disease frequency and association, bias assessment, and public health data analysis methods.
+
+## Study Design Selection
+
+### Design Hierarchy
+
+```
+                    Evidence Strength
+                         |
+    Systematic Review / Meta-Analysis   (Highest)
+                         |
+         Randomized Controlled Trial
+                         |
+              Cohort Study (Prospective)
+                         |
+            Case-Control Study
+                         |
+         Cross-Sectional Study
+                         |
+         Case Report / Case Series       (Lowest)
+```
+
+### When to Use Each Design
+
+| Design | Research Question | Time | Cost | Bias Risk |
+|--------|------------------|------|------|-----------|
+| RCT | Does intervention X prevent outcome Y? | Years | Very high | Lowest |
+| Prospective Cohort | Does exposure X increase risk of Y? | Years | High | Moderate |
+| Retrospective Cohort | Historical exposure-outcome relationship? | Months | Moderate | Moderate-High |
+| Case-Control | What exposures are associated with rare disease? | Months | Low | High |
+| Cross-Sectional | What is the prevalence of X? | Weeks | Low | High |
+| Ecological | Do population-level factors correlate with disease? | Weeks | Very low | Very high |
+
+## Measures of Disease Frequency
+
+```python
+import numpy as np
+
+def compute_measures(cases: int, population: int,
+                      person_time: float = None,
+                      period_years: float = 1.0) -> dict:
+    """
+    Compute basic epidemiological measures.
+
+    Args:
+        cases: Number of new cases (for incidence) or existing cases (for prevalence)
+        population: Population at risk
+        person_time: Person-years of follow-up (for incidence rate)
+        period_years: Time period in years (for cumulative incidence)
+    """
+    measures = {}
+
+    # Point prevalence
+    measures['prevalence'] = {
+        'value': cases / population,
+        'per_1000': (cases / population) * 1000,
+        'formula': 'cases / population at a point in time'
+    }
+
+    # Cumulative incidence (risk)
+    measures['cumulative_incidence'] = {
+        'value': cases / population,
+        'per_1000': (cases / population) * 1000,
+        'period_years': period_years,
+        'formula': 'new cases / population at risk during time period'
+    }
+
+    # Incidence rate (if person-time available)
+    if person_time:
+        measures['incidence_rate'] = {
+            'value': cases / person_time,
+            'per_1000_py': (cases / person_time) * 1000,
+            'formula': 'new cases / person-time at risk'
+        }
+
+    return measures
+```
+
+## Measures of Association
+
+### Risk Ratio, Odds Ratio, and Attributable Risk
+
+```python
+def measures_of_association(a: int, b: int, c: int, d: int) -> dict:
+    """
+    Compute epidemiological measures of association from a 2x2 table.
+
+                    Disease+    Disease-
+    Exposed+          a           b        a+b
+    Exposed-          c           d        c+d
+                     a+c         b+d        N
+
+    Args:
+        a: Exposed with disease
+        b: Exposed without disease
+        c: Unexposed with disease
+        d: Unexposed without disease
+    """
+    # Risk in exposed and unexposed
+    risk_exposed = a / (a + b)
+    risk_unexposed = c / (c + d)
+
+    # Risk Ratio (Relative Risk)
+    rr = risk_exposed / risk_unexposed
+    ln_rr = np.log(rr)
+    se_ln_rr = np.sqrt(1/a - 1/(a+b) + 1/c - 1/(c+d))
+    rr_ci = (np.exp(ln_rr - 1.96*se_ln_rr), np.exp(ln_rr + 1.96*se_ln_rr))
+
+    # Odds Ratio
+    or_val = (a * d) / (b * c)
+    ln_or = np.log(or_val)
+    se_ln_or = np.sqrt(1/a + 1/b + 1/c + 1/d)
+    or_ci = (np.exp(ln_or - 1.96*se_ln_or), np.exp(ln_or + 1.96*se_ln_or))
+
+    # Attributable Risk (Risk Difference)
+    ar = risk_exposed - risk_unexposed
+    se_ar = np.sqrt(risk_exposed*(1-risk_exposed)/(a+b) +
+                     risk_unexposed*(1-risk_unexposed)/(c+d))
+    ar_ci = (ar - 1.96*se_ar, ar + 1.96*se_ar)
+
+    # Attributable Fraction in Exposed
+    af_exposed = (rr - 1) / rr
+
+    # Population Attributable Fraction
+    prevalence_exposure = (a + b) / (a + b + c + d)
+    paf = prevalence_exposure * (rr - 1) / (prevalence_exposure * (rr - 1) + 1)
+
+    return {
+        'risk_ratio': {'value': round(rr, 3), 'ci_95': tuple(round(x, 3) for x in rr_ci)},
+        'odds_ratio': {'value': round(or_val, 3), 'ci_95': tuple(round(x, 3) for x in or_ci)},
+        'risk_difference': {'value': round(ar, 4), 'ci_95': tuple(round(x, 4) for x in ar_ci)},
+        'attributable_fraction_exposed': round(af_exposed, 3),
+        'population_attributable_fraction': round(paf, 3),
+        'number_needed_to_harm': round(1/ar, 1) if ar > 0 else None
+    }
+
+# Example: smoking and lung cancer
+result = measures_of_association(a=80, b=920, c=10, d=990)
+print(f"RR: {result['risk_ratio']['value']} ({result['risk_ratio']['ci_95']})")
+print(f"OR: {result['odds_ratio']['value']} ({result['odds_ratio']['ci_95']})")
+print(f"PAF: {result['population_attributable_fraction']}")
+```
+
+## Bias Assessment
+
+### Types of Bias and Mitigation
+
+| Bias Type | Description | Mitigation Strategy |
+|-----------|------------|-------------------|
+| Selection bias | Non-random sample selection | Random sampling, matching |
+| Information bias | Measurement error in exposure/outcome | Validated instruments, blinding |
+| Recall bias | Differential recall by disease status | Use records, not self-report |
+| Confounding | Third variable affects both exposure and outcome | Stratification, regression, matching |
+| Lead-time bias | Earlier detection misinterpreted as longer survival | Use mortality, not survival |
+| Healthy worker effect | Workers are healthier than general population | Use employed comparison group |
+
+### Confounding Assessment
+
+```python
+def assess_confounding(crude_rr: float, adjusted_rr: float,
+                        threshold: float = 0.10) -> dict:
+    """
+    Assess whether a variable is a confounder.
+    """
+    pct_change = abs(crude_rr - adjusted_rr) / crude_rr * 100
+
+    return {
+        'crude_RR': crude_rr,
+        'adjusted_RR': adjusted_rr,
+        'percent_change': round(pct_change, 1),
+        'is_confounder': pct_change > threshold * 100,
+        'interpretation': (
+            f"{'Confounder detected' if pct_change > threshold * 100 else 'Not a confounder'}: "
+            f"adjusting changed the RR by {pct_change:.1f}% "
+            f"(threshold: {threshold*100:.0f}%)"
+        )
+    }
+```
+
+## Survival Analysis
+
+For time-to-event data, use Kaplan-Meier estimators for descriptive analysis, log-rank tests for group comparisons, and Cox proportional hazards regression for multivariable analysis. Always check the proportional hazards assumption using Schoenfeld residuals and report median survival times with 95% confidence intervals.
+
+## Reporting Standards
+
+Follow STROBE (observational studies), CONSORT (trials), or RECORD (routinely collected data) reporting guidelines. Report all measures with 95% confidence intervals. Present both crude and adjusted estimates to show the impact of confounding adjustment.

package/skills/domains/biomedical/genomics-analysis-guide/SKILL.md

@@ -0,0 +1,270 @@
+---
+name: genomics-analysis-guide
+description: "Workflows for RNA-seq, GWAS, and variant calling in genomic research"
+metadata:
+  openclaw:
+    emoji: "microscope"
+    category: "domains"
+    subcategory: "biomedical"
+    keywords: ["genomics", "RNA-seq", "GWAS", "molecular biology", "genetics", "bioinformatics"]
+    source: "wentor-research-plugins"
+---
+
+# Genomics Analysis Guide
+
+## Overview
+
+Genomic data analysis is the computational backbone of modern molecular biology. From identifying disease-associated variants through Genome-Wide Association Studies (GWAS) to quantifying gene expression with RNA-seq, these workflows transform raw sequencing data into biological insights that drive discoveries in medicine, agriculture, and evolutionary biology.
+
+This guide covers the three most common genomic analysis workflows: RNA-seq differential expression analysis, GWAS for variant-trait associations, and variant calling from whole-genome sequencing (WGS) data. Each workflow is described with tool recommendations, command-line examples, and downstream analysis steps in R and Python.
+
+The emphasis is on reproducibility and best practices. Genomic analyses involve many sequential steps, and errors in early stages propagate through the entire pipeline. Following standardized workflows -- like those from the Broad Institute, ENCODE, and Bioconductor -- reduces the risk of methodological errors.
+
+## RNA-seq Analysis Pipeline
+
+### Workflow Overview
+
+```
+Raw FASTQ files
+    |
+    v
+[Quality Control] --> FastQC, MultiQC
+    |
+    v
+[Trimming] --> Trimmomatic, fastp
+    |
+    v
+[Alignment] --> STAR, HISAT2
+    |
+    v
+[Quantification] --> featureCounts, Salmon
+    |
+    v
+[Differential Expression] --> DESeq2, edgeR
+    |
+    v
+[Pathway Analysis] --> clusterProfiler, GSEA
+```
+
+### Step 1: Quality Control
+
+```bash
+# Run FastQC on all FASTQ files
+fastqc -t 8 -o qc_results/ raw_data/*.fastq.gz
+
+# Aggregate QC reports
+multiqc qc_results/ -o multiqc_report/
+```
+
+### Step 2: Read Trimming
+
+```bash
+# fastp for quality trimming and adapter removal
+fastp \
+  --in1 sample_R1.fastq.gz \
+  --in2 sample_R2.fastq.gz \
+  --out1 trimmed_R1.fastq.gz \
+  --out2 trimmed_R2.fastq.gz \
+  --detect_adapter_for_pe \
+  --thread 8 \
+  --html fastp_report.html
+```
+
+### Step 3: Alignment with STAR
+
+```bash
+# Build genome index (one time)
+STAR --runMode genomeGenerate \
+  --genomeDir star_index/ \
+  --genomeFastaFiles genome.fa \
+  --sjdbGTFfile annotations.gtf \
+  --runThreadN 16
+
+# Align reads
+STAR --runMode alignReads \
+  --genomeDir star_index/ \
+  --readFilesIn trimmed_R1.fastq.gz trimmed_R2.fastq.gz \
+  --readFilesCommand zcat \
+  --outSAMtype BAM SortedByCoordinate \
+  --quantMode GeneCounts \
+  --outFileNamePrefix sample_ \
+  --runThreadN 16
+```
+
+### Step 4: Differential Expression with DESeq2
+
+```r
+library(DESeq2)
+
+# Load count matrix and sample info
+counts <- read.csv("gene_counts.csv", row.names = 1)
+coldata <- read.csv("sample_info.csv", row.names = 1)
+
+# Create DESeq2 object
+dds <- DESeqDataSetFromMatrix(
+  countData = counts,
+  colData = coldata,
+  design = ~ condition
+)
+
+# Filter low-count genes
+keep <- rowSums(counts(dds) >= 10) >= 3
+dds <- dds[keep, ]
+
+# Run differential expression
+dds <- DESeq(dds)
+res <- results(dds, contrast = c("condition", "treated", "control"),
+               alpha = 0.05)
+
+# Summary
+summary(res)
+
+# Export significant genes
+sig_genes <- subset(as.data.frame(res), padj < 0.05 & abs(log2FoldChange) > 1)
+write.csv(sig_genes, "significant_genes.csv")
+```
+
+## GWAS Pipeline
+
+### Workflow Overview
+
+```
+Genotype Data (VCF/PLINK)
+    |
+    v
+[Quality Control] --> Sample/variant filtering
+    |
+    v
+[Population Stratification] --> PCA
+    |
+    v
+[Association Testing] --> PLINK2, REGENIE
+    |
+    v
+[Multiple Testing Correction] --> Bonferroni, FDR
+    |
+    v
+[Visualization] --> Manhattan plot, QQ plot
+```
+
+### QC with PLINK2
+
+```bash
+# Sample QC
+plink2 \
+  --bfile dataset \
+  --mind 0.05 \          # Remove samples with >5% missing
+  --geno 0.02 \          # Remove variants with >2% missing
+  --maf 0.01 \           # Remove rare variants (MAF < 1%)
+  --hwe 1e-6 \           # HWE filter
+  --make-bed \
+  --out dataset_qc
+
+# LD pruning for PCA
+plink2 \
+  --bfile dataset_qc \
+  --indep-pairwise 50 5 0.2 \
+  --out pruned
+
+# PCA for population stratification
+plink2 \
+  --bfile dataset_qc \
+  --extract pruned.prune.in \
+  --pca 10 \
+  --out pca_results
+```
+
+### Association Testing
+
+```bash
+# Linear/logistic regression with covariates
+plink2 \
+  --bfile dataset_qc \
+  --glm \
+  --pheno phenotypes.txt \
+  --covar pca_results.eigenvec \
+  --covar-col-nums 3-12 \
+  --out gwas_results
+```
+
+### Manhattan Plot in Python
+
+```python
+import pandas as pd
+import matplotlib.pyplot as plt
+import numpy as np
+
+def manhattan_plot(gwas_file, output='manhattan.pdf'):
+    df = pd.read_csv(gwas_file, sep='\t')
+    df['-log10p'] = -np.log10(df['P'])
+
+    # Assign cumulative positions
+    df = df.sort_values(['CHR', 'BP'])
+    df['pos_cum'] = 0
+    offset = 0
+    for chrom in df['CHR'].unique():
+        mask = df['CHR'] == chrom
+        df.loc[mask, 'pos_cum'] = df.loc[mask, 'BP'] + offset
+        offset = df.loc[mask, 'pos_cum'].max()
+
+    fig, ax = plt.subplots(figsize=(16, 5))
+    colors = ['#3B82F6', '#94A3B8']
+    for i, chrom in enumerate(df['CHR'].unique()):
+        subset = df[df['CHR'] == chrom]
+        ax.scatter(subset['pos_cum'], subset['-log10p'],
+                   s=2, color=colors[i % 2], alpha=0.7)
+
+    ax.axhline(-np.log10(5e-8), color='red', linestyle='--', linewidth=0.8)
+    ax.set_xlabel('Chromosome')
+    ax.set_ylabel('-log10(p-value)')
+    fig.savefig(output, dpi=300, bbox_inches='tight')
+```
+
+## Variant Calling Pipeline
+
+### GATK Best Practices
+
+```bash
+# Mark duplicates
+gatk MarkDuplicates \
+  -I aligned.bam \
+  -O dedup.bam \
+  -M metrics.txt
+
+# Base quality score recalibration
+gatk BaseRecalibrator \
+  -I dedup.bam \
+  -R reference.fa \
+  --known-sites dbsnp.vcf \
+  -O recal_table.txt
+
+gatk ApplyBQSR \
+  -I dedup.bam \
+  -R reference.fa \
+  --bqsr-recal-file recal_table.txt \
+  -O recal.bam
+
+# Call variants
+gatk HaplotypeCaller \
+  -I recal.bam \
+  -R reference.fa \
+  -O variants.g.vcf \
+  -ERC GVCF
+```
+
+## Best Practices
+
+- **Use containerized workflows.** Nextflow + Docker/Singularity ensures reproducibility across environments.
+- **Document every parameter.** Small changes in alignment settings can significantly affect downstream results.
+- **Apply appropriate multiple testing corrections.** Genome-wide significance is p < 5e-8 for GWAS.
+- **Validate findings in independent cohorts.** Replication is essential before biological interpretation.
+- **Archive raw data and analysis scripts.** Deposit in GEO (expression) or dbGaP (genotypes) for reproducibility.
+- **Use established pipelines (nf-core).** Community-maintained Nextflow pipelines encode best practices.
+
+## References
+
+- [DESeq2 Vignette](https://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html) -- Differential expression analysis
+- [GATK Best Practices](https://gatk.broadinstitute.org/hc/en-us/sections/360007226651-Best-Practices-Workflows) -- Variant calling
+- [PLINK2 Documentation](https://www.cog-genomics.org/plink/2.0/) -- Genetic association analysis
+- [nf-core Pipelines](https://nf-co.re/) -- Community Nextflow workflows
+- [RNA-seq Analysis Tutorial](https://rnabio.org/) -- Griffith Lab comprehensive tutorial

package/skills/domains/business/market-analysis-guide/SKILL.md

@@ -0,0 +1,112 @@
+---
+name: market-analysis-guide
+description: "Structured frameworks for market sizing, competitive analysis, and strategic ..."
+metadata:
+  openclaw:
+    emoji: "chart_with_upwards_trend"
+    category: "domains"
+    subcategory: "business"
+    keywords: ["market analysis", "strategic management", "operations management", "competitive analysis", "market sizing"]
+    source: "wentor"
+---
+
+# Market Analysis Guide
+
+A comprehensive skill for conducting rigorous market analysis in academic and applied research contexts. This guide covers quantitative market sizing, competitive landscape mapping, and strategic positioning frameworks grounded in peer-reviewed methodologies.
+
+## Market Sizing Methodologies
+
+Market sizing is the foundation of any credible market analysis. There are two primary approaches, and robust research typically employs both for triangulation.
+
+**Top-Down Approach (TAM/SAM/SOM)**
+
+Start with the total addressable market and narrow systematically:
+
+```
+TAM (Total Addressable Market)
+  -> SAM (Serviceable Available Market)
+    -> SOM (Serviceable Obtainable Market)
+
+Example calculation:
+  TAM = Global higher-education EdTech spend = $340B (2025, HolonIQ)
+  SAM = AI-powered research tools segment  = $12B
+  SOM = Realistic capture in Year 3        = $120M (1% of SAM)
+```
+
+**Bottom-Up Approach**
+
+Build estimates from unit economics:
+
+```python
+# Bottom-up market sizing
+users_in_target_segment = 8_000_000  # global PhD + postdoc researchers
+adoption_rate = 0.05                  # 5% in first 3 years
+avg_revenue_per_user = 180            # USD/year
+bottom_up_estimate = users_in_target_segment * adoption_rate * avg_revenue_per_user
+# Result: $72,000,000
+```
+
+Always cite the data sources for each assumption. Use government statistics (e.g., NSF, Eurostat), industry reports (Gartner, McKinsey), and published academic datasets.
+
+## Competitive Analysis Frameworks
+
+### Porter's Five Forces
+
+Apply Porter's framework systematically to map industry structure:
+
+| Force | Key Questions | Data Sources |
+|-------|--------------|--------------|
+| Rivalry | How many direct competitors? Market concentration (HHI)? | Crunchbase, SEC filings |
+| New Entrants | Capital requirements? Regulatory barriers? | Patent databases, regulatory filings |
+| Substitutes | What alternatives exist? Switching costs? | User surveys, app store data |
+| Buyer Power | Customer concentration? Price sensitivity? | Industry reports, interviews |
+| Supplier Power | Input scarcity? Vendor lock-in? | Supply chain databases |
+
+### SWOT and TOWS Matrix
+
+Go beyond basic SWOT by constructing a TOWS matrix that generates actionable strategies:
+
+```
+              Strengths (S)           Weaknesses (W)
+Opportunities  SO strategies           WO strategies
+  (O)          (use S to exploit O)    (overcome W via O)
+Threats        ST strategies           WT strategies
+  (T)          (use S to counter T)    (minimize W, avoid T)
+```
+
+## Data Collection and Validation
+
+Primary data collection methods for market analysis research:
+
+1. **Structured interviews** with industry experts (N >= 12 for saturation)
+2. **Survey instruments** validated with Cronbach's alpha >= 0.70
+3. **Conjoint analysis** for preference and willingness-to-pay estimation
+4. **Web scraping** of pricing pages, job postings, and product changelogs
+
+Secondary data sources to cross-validate:
+
+- Statista, IBISWorld, Grand View Research for market reports
+- USPTO/EPO patent filings for technology trajectory analysis
+- PitchBook/Crunchbase for funding and M&A activity
+
+## Reporting and Visualization
+
+Present findings using clear, reproducible visualizations:
+
+```python
+import matplotlib.pyplot as plt
+import numpy as np
+
+segments = ['Segment A', 'Segment B', 'Segment C', 'Segment D']
+sizes = [45, 28, 18, 9]
+colors = ['#3B82F6', '#EF4444', '#10B981', '#F59E0B']
+
+fig, ax = plt.subplots(figsize=(8, 6))
+ax.barh(segments, sizes, color=colors)
+ax.set_xlabel('Market Share (%)')
+ax.set_title('Competitive Landscape by Segment')
+plt.tight_layout()
+plt.savefig('market_share.png', dpi=300)
+```
+
+Always include confidence intervals or sensitivity ranges for quantitative estimates. A well-structured market analysis report should contain an executive summary, methodology section, findings with visualizations, and a limitations discussion.