Understanding significance of enrichR results

Question

[kathryn.eckartt]

I have a set of differential expressed genes from RNAseq that I'm running basic enrichment analysis on using enrichR.

How is one meant to interpret the combined score value in the resulting data frame? Should this be used to determine significance instead of the adjusted P value?

I thought it might be so given Figure 2 in the original enrichR paper (Chen et al, 2013) but I haven't taken a statistics course yet despite being a first year in a biomedical PhD program. Thank you for all of your help

Answer 1

[Charles Warden]

I haven't tested the Bioconductor version, but perhaps this message thread can be useful?

A: the best combine score for enrichment analysis from Enrichr

In terms of Enrichr interpretation, I usually check the first few categories ranked to show in the web-interface. However, I admittedly can't say for certain whether this is the "best" option. For IPA, sorting by either p-value or z-score can be helpful for different projects, so my guess is that use of the scores for Enrichr may be similar (meaning sometimes it is good to use the provided ranking, sometimes it is good to sort other ways).