Peak calling for multiple replicates with Genrich

Question

[linelr]

Hi

I´m working with ATAC data and have used genrich to call peaks for the replicates collectively.

Do genrich only report consensus peaks or does it merge peaks found in one or more replicates in addition to calling peaks that are present in only one of the replicates?

Thank you very much in advance

Kind regards, Line

[ATpoint]

Did you use Genrich recently, if so could you share your experiences, would be happy to hear how it performs in your hands. I tested it on a couple of samples (ATAC-seq) providing replicated data, and actually was quite happy with its performance. Still, this is all based on eye-balling on the IGV browser.

Answer 1

[Friederike]

Genrich calls peaks for multiple replicates collectively. First, it analyzes the replicates separately, with p-values calculated for each. At each genomic position, the multiple replicates' p-values are then combined by Fisher's method. The combined p-values are converted to q-values, and peaks are called. source

Does that address your question?