Dear all,
I'm very new in Microarray analysis, so I hope my question wouldn't be too irritating to you all.
I have pre-processed and filtered my microarray datasets and perform differential expression analysis. I would like to note that I do not consolidate my probes.
I have two probes detecting my interested gene, so I have 2 FDR values. I want to collapse them to acquire only 1 FDR value for my gene (and still keep the FDR value of each probe).
In brief, I want to combine two probe FDR values acquired from the same test in the same datasets (I hope I do not make any mistake here)
Would you please suggest me if this is possible and how to do it?
Thank you very much in advance,
Kaj
R version 3.2.4 (2016-03-10)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.4 LTS
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=th_TH.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=th_TH.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=th_TH.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=th_TH.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
Thank you very much for your response. First of all, I would like to apologize about sessionInfo(). I perform the differential test using 'lpe' package, and I corrected the FDR value by resampling FDR method.
The reason I need to combine the probe to the same gene is as indicated in another question of mine (that no one seems to be interested in it at all).
http://stats.stackexchange.com/questions/201520/combining-fdr-value-acquired-from-different-microarray-probes
As you can notice in my example, I have two probe giving different FDR for each paired test. I simply want to have another FDR value as overall representative value of the gene.
At this point, I'm going through probe consolidation as you suggest (so have another whole bunch analysis).
However, I have noticed that you mention about p-value as the least I required. As far as I can read (and poorly understand), I can only think of Fisher Exact test and Empirical Brown method. But none of them seems to fit my condition at all. Would you please suggest me or correct me (if I mistake something) about the test to combine p-value in my condition ?
Thank you very much in advance.
Kaj