Question: Pair-wise DESeq Analysis with 3 Factor Levels - RNA-Seq
0
gravatar for tasnif.rahman
11 months ago by
tasnif.rahman0 wrote:

Hello, very new to RNA-Sequencing, R and DESeq2 so the question might be too simple or repeated so I apologize in advance.

I am working with 24 RNA-Sequencing samples, 8 conditions with 3 biological replicates for each. A part of my data matrix is shown below.

enter image description here

Each sample represents one of three phenotypes. A wild-type phenotype, and two altered phenotypes. I would like to compare the wild-type phenotype with each of the other phenotypes, and the other two phenotypes with each other. I am confused as to what design formula I should use to generate my DESeq object. I have read the vignettes and beginner's guide but I lack the background to fully understand how the design formulas work. I have included part of my coldata below.

enter image description here

I would like to compare NB and CW, and NB and CCW. I aim to find genes that are differentially expressed as phenotype changes from NB to CW and from NB to CCW. Then I would like to compare CW and CCW. I would like to do this pair-wise for each sample pair and retain that information in the output. So far, I have been doing this by splitting my matrix into pairs of samples and running each pair individually but that is incredibly tedious and was wondering if I could accomplish the same results by using the complete expression matrix and coldata. I have tried only using chirality in the design formula (design = ~ Chirality) and set NB as the reference lever using the relevel() function, but upon using the resultsNames(dds) function I get,

"Intercept" "Chirality_CW_vs_CCW" "Chirality_NB_vs_CCW"

However, I believe I am looking for NB vs CW and NB vs CCW. Also, I have not included the Cell.Type in the design formula but I assume it should be included since I want to compare the samples pair-wise. Any help regarding design formulae for this particular case, understanding design formulae in R, in general, or other comments are greatly appreciated.

Thanks,

Tasnif Rahman

ADD COMMENTlink modified 11 months ago • written 11 months ago by tasnif.rahman0
0
gravatar for tasnif.rahman
11 months ago by
tasnif.rahman0 wrote:

The images have not posted in my original post. I have included them here.

countdata:

enter image description here

coldata:

enter image description here

ADD COMMENTlink modified 11 months ago • written 11 months ago by tasnif.rahman0

Please note the cross post on both sites:

https://support.bioconductor.org/p/115065/

ADD REPLYlink written 11 months ago by Michael Love1.9k
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: 796 users visited in the last hour