This repository houses modular workflows for statistical analyses of correlates of risk / protection and automated reporting of analytic results. It serves as a generalized suite of tools, based on the analyses originally designed for the USG Biostatistics Response Team's analysis of COVID-19 vaccine efficacy trials (archived here). See below for brief descriptions of each of the analysis modules. This repository is designed as the second part of an analytic pipeline, with the correlates processing module serving as an upstream component.

Note: automated builds of the correlates of risk and protection analyses are evaluated by the Travis CI continuous integration service and the PDF reports posted to this repository's gh-pages branch.

• Correlates of Risk (CoR) Analyses
  • cor_tabular: Tabular descriptions of correlates of risk.
  • cor_graphical: Graphical descriptions of correlates of risk.
  • cor_coxph: Cox proportional hazards modeling of risk.
  • cor_threshold: Risk modeling based on correlate thresholds.
  • cor_nonlinear: Nonlinear modeling and evaluation.
  • cor_surrogates: Optimal surrogates analyses.
• Correlates of Protection (CoP) Analyses
  • cop_prinstrat: Principal stratification analyses.
  • cop_stochastic: Stochastic risk and vaccine efficacy evaluation.
  • cop_mediation: Correlate-mediated vaccine efficacy and risk.

• Code style guide, with some modifications; this will largely be enforcd with styler.
• Project organization: mostly independent subdirectories, each incorporating here for path resolution.
• Package version control and virtual environments using renv.
• Code review procedure: see our contribution guidelines.

The contents of this repository are distributed under the GPL-3 license. See file LICENSE.md for details.