Question: DESeq2 for ChIP on chip microarrays
gravatar for Assa Yeroslaviz
11 days ago by
Assa Yeroslaviz1.3k
Munich, Germany
Assa Yeroslaviz1.3k wrote:

I would like to know if it is possible to use DESeq2 to analyze NimbleGene microarrays. I have a dataset of several conditions with enriched (IP) and control (Input) sample for several knock-outs. We would like to try and identify enriched probes (= this is what equals the gene name on the microarray) both within an IP-Input comparison, but also between two conditions.

a short example - These is one of the comparison I would like to run:


file condition assay
561534A07_532.pair 561534A07_532 WT Input Y0003_FK2
561534A07_635.pair 561534A07_635 WT IP Y0003_FK2
561500A01_532.pair 561500A01_532 WT IP Y0003_FK2_2
561500A01_635.pair 561500A01_635 WT Input Y0003_FK2_2
561565A01_532.pair 561565A01_532 cdc48 Input MJK503_FK2_1
561565A01_635.pair 561565A01_635 cdc48 IP MJK503_FK2_1
561565A10_532.pair 561565A10_532 cdc48 IP MJK503_FK2_2
561565A10_635.pair 561565A10_635 cdc48 Input MJK503_FK2_2

My first comparison would be between the IP and Input of the WT and of the cdc48 sets separately, But I would also like to try and identify enriched regions/probes between samples.  this would be based on ~condition for the design formula.

But Is it also possible to calculate the ratio of ratios and compare the two data sets with each other?

What I would like to do is something like (IP of cdc48 / Input of cdc48) / (IP of WT / Input of WT)

should this be something like that:

dds <- DESeqDataSet(se, design= ~ assay + condition + assay:condition)
dds <- DESeq(dds, test="LRT", reduced= ~ assay + condition)

Is it possible to do something like that with DESeq2? I know the probe intensities are not counts, but maybe it is possible to read them into DESeq2 with the tximport package which than could convert them into counts.

thanks, Assa

ADD COMMENTlink modified 11 days ago by Michael Love14k • written 11 days ago by Assa Yeroslaviz1.3k
gravatar for Michael Love
11 days ago by
Michael Love14k
United States
Michael Love14k wrote:

DESeq2 is only for count data. For array data you should use a package designed to support array data, such as limma.

ADD COMMENTlink written 11 days ago by Michael Love14k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 180 users visited in the last hour