Question

deseq2 low read input

0

Entering edit mode

Bob • 0

@bob-12005

Last seen 6.2 years ago

Hi all

for prevent p-values set to NA in DESeq2, should I offset my raw reads as same as this ?

Thanks

      cont1  cont2 cont3 tumor1 tumor2 tumor3
geneA    64     80    36      0      0      0
geneB    66     87    56      1      0      0

deseq2 • 889 views

ADD COMMENT • link updated 7.2 years ago by Michael Love 43k • written 7.2 years ago by Bob • 0

0

Entering edit mode

What do you mean by "offset"?

ADD REPLY • link 7.2 years ago Simon Anders ★ 3.8k

0

Entering edit mode

replace 0 with 1 and add 1 to each control reads for instance change the raw reads to :

  cont1  cont2 cont3 tumor1 tumor2 tumor3
geneA    65     81    37      1      1      1
geneB    66     88    57      1      1      1

ADD REPLY • link 7.2 years ago Bob • 0

score 0 · Answer 1 · 2017-10-19

0

Entering edit mode

Michael Love 43k

@mikelove

Last seen 5 hours ago

United States

No you should not ever alter the counts for DESeq2. If you want to stop p-values being set to NA, see this section:

https://bioconductor.org/packages/release/bioc/vignettes/DESeq2/inst/doc/DESeq2.html#how-can-i-get-unfiltered-deseq2-results

ADD COMMENT • link 7.2 years ago Michael Love 43k

0

Entering edit mode

for gene A the basemean is 0 so p-value adjusted to NA and this gene shows dysregulation in tumor samples if I exclude some genes as same as this from count matrix I will have false positive/negative result, is it true?

ADD REPLY • link 7.2 years ago Bob • 0

0

Entering edit mode

Base mean of 0 means the experiment had no observations of the gene (according to how you are quantifying). DESeq2 can’t help you measure DE for a gene with no observations obviously.

ADD REPLY • link 7.2 years ago Michael Love 43k