[BioC] Low number of replicates DESeq

Federico Gaiti f.gaiti at uq.edu.au
Sat Mar 22 11:36:56 CET 2014

Thanks a lot. 
I spent a couple of days trying to figure how to make this. As usual quick reply and efficient!


From: mailinglist.honeypot at gmail.com [mailinglist.honeypot at gmail.com] on behalf of Steve Lianoglou [lianoglou.steve at gene.com]
Sent: Saturday, 22 March 2014 3:32 AM
To: Michael Love
Cc: Federico Gaiti; bioconductor at r-project.org
Subject: Re: [BioC] Low number of replicates DESeq


On Fri, Mar 21, 2014 at 4:48 AM, Michael Love
<michaelisaiahlove at gmail.com> wrote:
> hi Federico,
> See ?grep
> http://stat.ethz.ch/R-manual/R-devel/library/base/html/grep.html

To provide a little more help to your question. I'll take the code you
provided and modify it a bit:

I'd first extract all the results ordered by pvalue (not just the top
N as you previously did):

R> E <- assay(rld)[order(res$padj), ]

`rownames(E)` should have the gene/transcript IDs you want to grep
against. Then you could just filter out the ones with the "TCONS_"
pattern, ie:

R> E.tcons <- E[grepl('^TCONS_', rownames(E)), ]

Use the top N in your heatmap:

R> heatmap.2(E.tcons[1:N, ], ...)



PS: Doing a little email-thread pruning never hurt nobody! :-)

[15 pages of quoted email thread axed here]

Steve Lianoglou
Computational Biologist

More information about the Bioconductor mailing list