search

buenrostrolab / FigR

Functional Inference of Gene Regulation
https://buenrostrolab.github.io/FigR/
MIT License
31 stars 10 forks source link

Bulk ATAC and RNAseq data #43

Open lucalvizi opened 3 months ago

lucalvizi commented 3 months ago

Hi, I have some some bulk ATAC and RNA seq in which I wanted to identify some DORCs. Is it possible to use the dorc function in this kind os data (ie importing my bulk rds into the objects)? Of course not for UMAP analysis. Many thanks

best Lucas

vkartha commented 1 week ago

Hi there! We have actually run this code for bulk data as well - it works since it is based on sample / cell wise correlations between peak / gene expression counts, and will utilize background calculations the same way it would do for single cell data, but using bulk ATAC data. I'd say worth trying as long as your samples span different conditions (or contain enough variation) where it makes conceptual sense to perform a correlation between the ATAC/RNA across them (for e.g. if they were technical replicates of the same condition, this would not make much sense to do). The main difference in implementation here is that we would require you to normalize the data appropriately first due to the different sparsity / scale of the ATAC/RNA data compared to the single cell equivalent (e.g. , and then set normalizeATACmat to FALSE when running the function: https://buenrostrolab.github.io/FigR/reference/runGenePeakcorr.html). For e.g., DESeq2 normalized counts for RNAseq, and quantile normalized counts for bulk ATACseq.