SEQanswers

Go Back   SEQanswers > Bioinformatics > Bioinformatics



Similar Threads
Thread Thread Starter Forum Replies Last Post
edgeR ECHo Bioinformatics 13 04-18-2013 01:01 PM
Help the rookie with how to use edgeR- please! finnola Bioinformatics 1 11-08-2011 09:18 AM
edgeR p-value and logFC alessandra85 Bioinformatics 3 08-29-2011 05:04 AM
edgeR Puva Bioinformatics 2 05-19-2011 09:04 AM
DEGseq or EdgeR MerFer Bioinformatics 3 02-25-2010 12:48 AM

Reply
 
Thread Tools
Old 02-03-2012, 04:14 PM   #1
polsum
Member
 
Location: Texas

Join Date: May 2009
Posts: 32
Default EdgeR: Heatmaps

Hi, I know DESeq can make some really nice heat maps along with clustering of the differentially expressed genes. How can I make such maps with edger? Can I import edgeR data to Deseq to make those graphs?

Thanks in advance.
polsum is offline   Reply With Quote
Old 02-03-2012, 04:20 PM   #2
chadn737
Senior Member
 
Location: US

Join Date: Jan 2009
Posts: 392
Default

The commands used to make heatmaps in the DESeq vignette are not part of the DESeq package. You can make them easily enough in R using any data. You will just have to modify them to recognize the appropriate fields in EdgeR.
chadn737 is offline   Reply With Quote
Old 02-04-2012, 12:40 AM   #3
Simon Anders
Senior Member
 
Location: Heidelberg, Germany

Join Date: Feb 2010
Posts: 994
Default

However, what is part of DESeq is the function 'getVarianceStabilizedData' that transforms the data to a scale suitable as input to clustering or heatmap functions.
Simon Anders is offline   Reply With Quote
Old 02-04-2012, 04:43 AM   #4
gringer
David Eccles (gringer)
 
Location: Wellington, New Zealand

Join Date: May 2011
Posts: 838
Default

Quote:
Originally Posted by Simon Anders View Post
However, what is part of DESeq is the function 'getVarianceStabilizedData' that transforms the data to a scale suitable as input to clustering or heatmap functions.
would it be reasonable to filter out transcripts / genes that have low read counts prior to doing this clustering (e.g. fewer than 5 reads in all samples, or fewer than 50 reads in any sample)? Alternaltively, will clustering with the variance-stabilized data work better if you use only the most expressed transcripts (e.g. top 1000, based on rowSums)?
gringer is offline   Reply With Quote
Old 02-04-2012, 05:49 AM   #5
Simon Anders
Senior Member
 
Location: Heidelberg, Germany

Join Date: Feb 2010
Posts: 994
Default

That depends a lot on what you hope to achieve with the clustering. A classical application is to see similarity of samples; and in that case you would not want to restrict yourself to only the most strongly expressed genes. However, as clustering with all genes is often computationally to expensive, it may help to kick out low-variance genes, which have the same value in all samples and so help little to distinguish them.
Simon Anders is offline   Reply With Quote
Old 02-04-2012, 11:47 AM   #6
polsum
Member
 
Location: Texas

Join Date: May 2009
Posts: 32
Default

Quote:
Originally Posted by Simon Anders View Post
However, what is part of DESeq is the function 'getVarianceStabilizedData' that transforms the data to a scale suitable as input to clustering or heatmap functions.
Simon, this precisely is my question. The function 'getVarianceStabilizedData' uses 'cds' as the argument which doesnt have same structure as argument 'd' generated by EdgeR. My question is how to give 'd' as the argument to this function?

Sorry, I am very naive to 'R' language.

Thanks
polsum is offline   Reply With Quote
Reply

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 06:19 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2020, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO