Analysing two multiome data sets (RNA + ATAC)

[borjaarozamena]

Hi,

I am new in analysing single cell sequencing data and I have some doubts about the correct form of processing the data. I performed a multiome analysis of a population of cells for control and tretament. I processed everything together and sequence together but in different lines to differentiate, so I have both data independently but I dont think I have batch effect.

My doubts come at the moment on starting with the ATAC data. I have seen that is good to redo the calling peaks using the Macs2 but after, as I am doing it independently, I dont know if I have to do an integration of the peaks or follow the merging vignette and create a common data set peaks. Here, I dont know if creating a common data set just taking the peaks of one dataset, I will lose peaks of the other dataset that are not represent in the first one. Also, Is it possible to run this using the h5 file or fragments file instead of bed files. As I mention before I dont think I have this batch effect so for the RNA and ATAC is only necessary to merge the data and not an integration?

Also, I have found some issues about the function FindMarkers() for the differential accessible peaks between conditions in the same cluster. Because when I tried with the data (also I dont know if the upstream processing of the data is correct), I found when plotting using the CoveragePlot() function, sometimes it is not corresponding as I expected. So, I am thinking the issue that came up in #163 and if there is some problems with this function for peaks and maybe I have to work with something different form the avg_log2FC.

Thank you in advance,
Borja.