Go Back   SEQanswers > Bioinformatics > Bioinformatics

Similar Threads
Thread Thread Starter Forum Replies Last Post
DEXSeq with more than one condition unicornich Bioinformatics 1 03-21-2016 04:13 AM
DESeq2 and zero counts in one condition gmatt RNA Sequencing 1 01-09-2015 11:32 AM
Paired-test in DESeq or DESeq2 gcai Bioinformatics 3 06-17-2014 02:57 AM
Paired design versus unpaired design in DESeq2 KristenC RNA Sequencing 1 05-29-2014 12:05 PM
DESeq2: Difference between condition+type vs. 3 conditions mevers Bioinformatics 3 09-02-2013 02:17 AM

Thread Tools
Old 02-23-2015, 01:00 PM   #1
Location: Virginia

Join Date: Mar 2011
Posts: 72
Default DESeq2 paired plus condition


I am having some trouble getting my head around it today, its Monday.

Essentially, I have a design matrix like so:

Patient Condition
58 2
58 3
64 2
64 3

So, according the the vignette, I need to do ~Patient + Condition. I really want condition differences, NOT patient differences.

I can view the results via results(slim_dds_1, contrast=c("Condition","3","2")). How do I know what was a patient difference? I suppose I could do the reduced formula and compare?

Last edited by bioBob; 02-23-2015 at 01:29 PM.
bioBob is offline   Reply With Quote
Old 02-23-2015, 01:28 PM   #2
Devon Ryan
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,480

You're going to need to provide rather more detail.
dpryan is offline   Reply With Quote
Old 02-24-2015, 08:09 AM   #3
Senior Member
Location: California

Join Date: Jul 2014
Posts: 198

You can extract patient differences like so:
results(slim_dds_1, contrast=c("Patient", "58", "64")
Specifying the design as you did will explicitly account for differences in Patient when looking at Condition.
fanli is offline   Reply With Quote
Old 02-24-2015, 08:23 AM   #4
Devon Ryan
Location: Freiburg, Germany

Join Date: Jul 2011
Posts: 3,480

@bioBob: for future reference, we don't get easily notified when a post gets edited. For a speedier reply, post a comment after editing (or just post the edit as a new comment). Anyway, I suspect that fanli answered your question. If instead you really want to see what genes vary with patient in general (i.e., not between any individual pair of patients but generally across them), then you can use the LRT test with a reduced model of ~Condition.
dpryan is offline   Reply With Quote

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

All times are GMT -8. The time now is 02:18 PM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO