cit

cit performs causal inference testing by providing P-values and permutation-based false discovery rate (FDR) q-values to evaluate potential mediators and causal relationships in biological datasets.

Key Features:

• Assumption testing: Implements hypothesis tests to directly evaluate the assumptions necessary for causal inference.
• P-value calculation: Computes P-values for potential mediators.
• Permutation-based FDR (q-values): Optionally estimates false discovery rates (q-values) using a non-parametric permutation approach.
• Variable and covariate flexibility: Handles single and multiple binary or continuous instrumental variables, binary or continuous outcome variables, and adjustment covariates.
• Type I error control: Uses permutation-based estimation to mitigate type I errors in multiple testing contexts.
• Simulation validation: Validation via simulation studies demonstrating effectiveness and robustness.
• Implementation: Includes embedded C++ code that leverages the GNU Scientific Library for computational efficiency.

Scientific Applications:

• Causal relationship assessment: Assess causal relationships and mediation effects in complex biological datasets.
• Mediator prioritization: Identify and prioritize potential mediators using P-values and permutation-derived q-values.
• Analysis in low-power or complex studies: Apply non-parametric permutation testing in studies with limited statistical power or complex data structures.

Methodology:

Performs hypothesis tests to produce P-values, optionally computes permutation-based FDR q-values via non-parametric permutations, employs simulation studies for validation, and implements computations in embedded C++ using the GNU Scientific Library.