Hi. I have whole exome raw count data for a bunch of samples, and I'm looking to compare gene expression for a single gene between samples, e.g., between samples that have some type of mutation in said gene and samples that don't. I'd like to do this analysis over a set of genes, but each one individually (that is, I only care about how a mutation in gene X affects expression in gene X, but not how it might affect expression elsewhere). The issue is, I'm not sure how to do this in DESeq2, which seems to be the most popular gene expression package for R. The main DEseq2 function requires a design in which each sample is assigned a condition (e.g., treated vs. untreated) and then applies its negative binomial model to the unnormalized counts, but it would be prohibitively time-consuming to rerun DESeq2 for each gene, using each gene's mutation status as the condition for each run.
So, what would be the best way to do just normalize my count data so that I can compare individual genes (either with DESeq2 or another package)? Also, any comments on what the optimal statistical approach would be for comparing expression between two groups of samples for an individual gene are also appreciated. Thanks.
