Hi There,
this is my first post so let's see how it goes.
I am new to transcriptomics and i am analysing my lab data to find differentially expressed genes. Currently, I have applied MTC using the "BH" method in LIMMA. However, in the step following the identification of DE genes i am intersecting DE genes from different contrasts to find a pool of DE genes common to all contrasts and continue with my analysis.
My question is the following, am i better off using no MTC and increase the number of false positives across contrasts since after i am intersecting 5 groups to isolate my pool of interesting genes, or shall i keep the MTC = BH method to control for FDR and therefore have a smaller number of genes to start with?
I am interested to know someone's opinion.
Christian
this is my first post so let's see how it goes.
I am new to transcriptomics and i am analysing my lab data to find differentially expressed genes. Currently, I have applied MTC using the "BH" method in LIMMA. However, in the step following the identification of DE genes i am intersecting DE genes from different contrasts to find a pool of DE genes common to all contrasts and continue with my analysis.
My question is the following, am i better off using no MTC and increase the number of false positives across contrasts since after i am intersecting 5 groups to isolate my pool of interesting genes, or shall i keep the MTC = BH method to control for FDR and therefore have a smaller number of genes to start with?
I am interested to know someone's opinion.
Christian