search

RegulatoryGenomicsGroup / chicdiff

A differential caller for capture Hi-C data
4 stars 3 forks source link

ChicDiff differential loop calling #18

Open FN2023 opened 1 year ago

FN2023 commented 1 year ago

Hi Mikhail

Your paper published in cell report 2020 sates in Method section that differential interaction calling was performed on merged-replicate data for two conditions.

Just to make sure if i have understood it right

1: Did you use "chicdiffPipeline" with peakMatrix created using all 4 samples including both conditions (one file) or peakMatrix per condition (two files) or peakMatrix per sample (4 files)?

2: and there were two .Rds files (one per condition) after runChicago.R ?

Thanks a lot

mspivakov commented 11 months ago

Sorry for the huge delay in replying, for some reason I wasn't getting notifications! Individual replicate Rds and chinput files should always be submitted to Chicdiff. However, the features (peaks) can be defined using either merged or individual Chicago calls. So (1) is correct but if I got you correctly (2) is not. I should also note that we used a developmental version of chicdiff in that paper, so I'm not sure the results obtained using the current version will be exactly the same - but they should be close!