bio-workflows-clinical-trial-pipeline - End-to-End Clinical Trial Analysis Pipeline
Analyzes clinical trial data from CDISC loading and preparation through estimand-driven primary, subgroup, sensitivity, and regulatory reporting analyses.
Tags
Updated: 2026-10-19Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Load CDISC domain files
- Build subject-level datasets
- Define estimands before analysis
- Run primary logistic models
- Perform categorical tests
- Analyze treatment subgroups
- Run missing-data sensitivity analyses
- Adjust graphical multiplicity
- Conduct survival analyses
- Generate baseline tables
- Produce regulatory reports
Inputs
- CDISC SDTM or ADaM domain files
- Statistical analysis plan
- Estimand specifications
- Treatment and endpoint definitions
- Intercurrent-event strategy
- Analysis population definitions
- Covariate specifications
- Subgroup definitions
Outputs
- Subject-level analysis dataset
- Primary analysis estimates
- Categorical test results
- Subgroup interaction results
- Missing-data sensitivity results
- Baseline Table 1
- Forest plots
- Regulatory tables and figures
Requirements
- Python environment
- statsmodels 0.14+
- scipy 1.12+
- tableone 0.9+
- pyreadstat 1.2+
- pandas 2.1+
- numpy 1.26+
- matplotlib 3.8+