M-value or P value? or both of them for microarray data analysis

Question

[eisa_kohan]

Dear all,

I am using microarray data for my studies and I wanna compare deferentially expressed genes across arrays. I selected genes based on M value more than 1 or less than -1 (means at least 2 fold changes in expression).

Now, Is it enough to define/list deferentially expressed genes? or I should calculate p value and select genes based on that? or I should use both of them (p value and m value; genes with p value less than 0.05 which has M value more than 1 or less than -1) to select genes? Thanks in advance for your answers.

Answer 1

[Devon Ryan]

You absolutely need to use the adjusted (not raw!) p-values. One will often filter the results by fold change (M) as well, since you're unlikely to want to follow up on small changes.