INTACT (INtegration of TWAS And ColocalizaTion) is an R package for constraining transcriptome-wide association study correlation statistics through an empirical Bayes approach. As input, INTACT takes gene-level colocalization probabilities and TWAS z-statistics, returning a posterior probability of putative causality for each gene.
The package also performs gene set enrichment estimation using probabilistic INTACT results and a list of pre-defined gene sets (INTACT-GSE).
For a thorough description of the methodology, refer to https://doi.org/10.1016/j.ajhg.2022.12.002.
library(devtools)
devtools::install_github("jokamoto97/INTACT")
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("INTACT")
Here is an example of how to integrate colocalization probabilities and TWAS
z-statistics for a simulated data set simdat:
intact(GLCP_vec=simdat$GLCP, z_vec = simdat$TWAS_z)
To perform gene set enrichment analysis using simdat and pre-defined gene
set list gene_set_list, run
intactGSE(gene_data = simdat, gene_sets = gene_set_list)
Please contact [email protected] or [email protected] if you have any questions.