Biostar Beta. Not for public use.
Question: Any paper or journal review about repeated multiple testing for ChIP-seq replicates?
1
Entering edit mode

Dear all:

I need to understand multiple repeated testing for biological ChIP-seq replicates, to observe test variation in repeated multiple test. Can anyone tell me any related papers, journal review to explain the topic that I raised? I also prefer to look at any existing Bioconductor packages that implement multiple repeated testing for ChIP-seq replicates. Thanks a lot.

Best regards:

Jura

Entering edit mode
1

What exactly do you mean by "multiple repeated testing"? Do you just mean "multiple testing", or do you mean some sort of iterative multiple testing issue?

ADD REPLYlink 3.6 years ago
Devon Ryan
90k
Entering edit mode
0

Dear Devon Ryan:

Thanks for your quick respond. I am looking for paper or journal review about Multiple testing for ChIP-seq replicates. Yes, I'll be grateful if you are able to interpret difference between multiple testing and iterative multiple testing that you mentioned. Thanks a lot

Best regards:

Jurat

ADD REPLYlink 3.6 years ago
Jurat Shahidin
• 40
Entering edit mode
1

Just read over the diffBind vignette for testing replicates. I'm not sure that they ever bothered writing a paper, but that's the standard tool.

Regarding some sort of iterative process, I've never heard of anyone needing to do that.

ADD REPLYlink 3.6 years ago
Devon Ryan
90k
Entering edit mode
1

I would take a look at this paper that came up with a method for calculating an "irreproducible discovery rate" (IDR), which they claim is analogous to the FDR but leverages replicates to quantify reproducibility. The method seems to particularly work well for ChiP-seq replicates, as that is the application they discuss. They also created the "idr" R package to calculate IDR which is found here (https://cran.r-project.org/web/packages/idr/index.html ). The method apparently works on all the replicates at once, so there is no need to iterate through each replicate and then some how combine them.

ADD REPLYlink 3.6 years ago
Collin
• 660

Login before adding your answer.

Similar Posts
Loading Similar Posts
Powered by the version 2.0