PCxN

Version submitted to PLOS Computational Biology

The scripts to get the pathway correlation estimates are in the src folder. Below is a brief overview of the scripts.

1. Update gene set annotations: Keep only genes present in the gene expression background. (mean_pcor2_barcode_hallmark_estimates00.R)

2. Get experiment-level estimates: For each experiment, estimate all pairwise pathway correlation coefficients along with the corresponding p-values. (mean_pcor2_barcode_hallmark_estimates01.R, mean_pcor2_barcode_hallmark_estimates01.sh)

3. Combine experiment-level estimates: Aggregate the experiment-level correlation estimates using a weighted average, and the experiment-level p-values using Liptak's method. (mean_pcor2_barcode_hallmark_estimates02.R, mean_pcor2_barcode_hallmark_estimates02.sh)

4. Aggregate results into a single data frame (mean_pcor2_barcode_hallmark_estimates03.R)