I'm using data from GEO and the GEO2R online analysis tool. The GEO2R analysis give adjusted P, P, t scores. I have 5 studies using the GEO2R tool. Is there a way to combine all these scores and filter out only significant genes?
A simple analysis is to use standard meta analysis tools, e.g., Fisher's test (see https://en.wikipedia.org/wiki/Fisher%27s_method).
However, keep in mind that this method may produce a low p-value even for genes that are differential in at least one of the studies. Nevertheless, it provides a useful ranking of the genes.
meta-analysis is right way t go. And fisher test for p value calculation is the easiest option.
there are few hidden problems in this analyses related do the direction of gene regulation and heterogeneity between studies that should be taken into account.
In our recent review we have described the process step by step .
I hope it should be easy to follow.
it is available here:
http://www.la-press.com/reverse-engeneering-of-regulatory-networks-from-big-data-a-roadmap-for-article-a4805
http://www.la-press.com/reverse-engeneering-of-regulatory-networks-from-big-data-a-roadmap-for-article-a4805
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools