Filtering Steps for proActiv Analysis

[lucianhu]

Dear GoekeLab,

I am reaching out with questions regarding the filtering step prior to running the ProActiv tool. In your paper, in the "Identification of tissue-specific alternative promoters" section, you mentioned that "we filtered out inactive genes by enforcing the mean gene expression ≥ 1." From my understanding, this means that before running getAlternativePromoters, I should first filter my results in the SummarizedExperiment object. Will this pre-filtering step affect the calculation results in any way?

Additionally, you mention adjusting the mean of absolute and relative promoter activity to be above ≥ 0.25. Does this mean we should filter each promoter's mean absolute and relative activity, or is it the mean promoter activity across all samples?

Thank you for your clarification on these filtering steps.

Have a good morning,

Nhu

[complexgenome]

Hi @lucianhu Nhu
Sorry to hijack your thread. I've a question in the same vein.

Is 0.25 a mean value of gene expression?