SEQanswers

SEQanswers (http://seqanswers.com/forums/index.php)
-   Bioinformatics (http://seqanswers.com/forums/forumdisplay.php?f=18)
-   -   What happened to per-condition and gene-est-only in DESeq2? (http://seqanswers.com/forums/showthread.php?t=61284)

friducha 07-16-2015 04:03 PM

What happened to per-condition and gene-est-only in DESeq2?
 
In DESeq, there was an option to estimate the gene dispersions within each condition instead of across all conditions, and per gene instead of sharing dispersion estimates across genes.

However, in DESeq2, I can't find these options anywhere -- how to specify these tweaks to the dispersion estimates now?

Thank you!

Michael Love 07-17-2015 07:26 AM

In DESeq2 we only estimate a single dispersion value per gene. See the methods description in the paper:

http://genomebiology.com/2014/15/12/550

friducha 07-17-2015 11:46 AM

Thanks, Michael.

So it is clear to me now that the dispersion values are now per-gene, so gene-est-only no longer applies.

Quoting from the paper:

Quote:

Many methods for differential expression analysis of RNA-seq data perform such information sharing across genes for variance (or, equivalently, dispersion) estimation. edgeR[2],[3] moderates the dispersion estimate for each gene toward a common estimate across all genes, or toward a local estimate from genes with similar expression strength, using a weighted conditional likelihood. Our DESeq method [4] detects and corrects dispersion estimates that are too low through modeling of the dependence of the dispersion on the average expression strength over all samples.

What about estimating the dispersions within each condition instead of across all samples? (per-condition)?

Thanks.


All times are GMT -8. The time now is 03:00 PM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2021, vBulletin Solutions, Inc.