Question: Comparing two RNA-seq datasets of same species and treatments
1
gravatar for sakuraazalea
12 months ago by
sakuraazalea10
sakuraazalea10 wrote:

I have RNA-seq studies from two labs that did similar experiments (used same species and both used the same factor with two labels - drug and control). One study had 3 replicates per label (6 samples all together) and other had 4 replicates per label (8 samples all together).

I am wondering if there is a way to compare the results between these two studies?

1) One method may be to compare the DEGs between the two studies. Say the first study had 20 DEGs and second study had 130 DEGs with 8 overlapping between the two studies. Is there a statistical method to test how similar these DEGs are in this situation?

2) Another method is to determine how similar the results from the two studies are not only looking at DEGs, but looking at ALL genes. Possibly determining how similar the FDR values for each and every gene is between the two studies? Maybe some ranking test of FDR values?

My main goal is to determine whether the two studies produced similar results regarding significance. Any advice is greatly appreciated.

significance rna-seq • 433 views
ADD COMMENTlink modified 12 months ago by Sean Davis25k • written 12 months ago by sakuraazalea10
0
gravatar for Sean Davis
12 months ago by
Sean Davis25k
National Institutes of Health, Bethesda, MD
Sean Davis25k wrote:

Using a GSEA approach where one study's DEG list is the "gene set" which you can test for enrichment in the other will give you a statistical result. I think this is one answer for both of your questions.

There are other ways to look at individual gene-level results, including meta-analysis.

ADD COMMENTlink written 12 months ago by Sean Davis25k

Thank you. One of the data had only ~40 DEGs and the other had ~1000 DEGs. So, you are suggesting to put, say, the 1000 DEGs from the second study as the "gene set" in a GSEA software and calculate enrichment of these 1000 DEGs in the first study (that only had 40 DEGs)? In this scenario, shouldn't I expect it to not be significant since I already know that the first study only had 40 DEGs? Or maybe that would not matter?

Do you have any suggestions on software to look into meta-analysis of individual gene-levels results between two studies of same species?

Thank you again for your sharing you advice!

ADD REPLYlink written 12 months ago by sakuraazalea10
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2519 users visited in the last hour