pvalue or adj_pvalue

[Flu09]

I used findmarkers() in Seurat to find DEGs I wonder which is the right metric to use pvalue or adj_pvalue

[humanpaingeneticslab]

For input to 'CreateiDEAObject', you will need neither of those. Rather, it takes the log2 fold changes and their variances.

Generally speaking, if you report the results of Seurat for DEGs, then please use the 'adj_pvalue'. When reporting on multiple statistical tests (in your case one test per gene) then you want to use the adjusted P-values (typically FDR-corrected, as genes aren't independent of one another, so Bonferroni's correction for multiple testing would be too pessimistic).