JiayiJi/iProMix: Cell-Type Specific Association Analysis for Bulk Profiling Data

Human tissues are comprised of multiple cell types with varying compositions. The bulk-tissue profiles of multiple -omic data types (e.g. DNA methylation, mRNA, proteomics) are impacted by the cell-type composition heterogeneity, as their levels in different cell types may be different. 'iProMix' decomposes data from single/multiple -omic data types (e.g. DNA methylation, mRNA, proteomics) and evaluates their cell-type specific dependences. A major difference of 'iProMix' from previous studies is that it allows association analysis on two data types that are both affected by cell types (e.g. mRNA vs. protein). It builds in features to improve cell type composition estimation if existing estimates are not satisfactory. It also takes into consideration the effects of decomposition and biased input on hypothesis tests, and generates valid inference in non-asymptotic settings.

install.packages("remotes")
remotes::install_github("JiayiJi/iProMix")