Search
Question: DESeq2 negative tag counts
0
gravatar for xiaofeiwang198266
14 months ago by
xiaofeiwang1982660 wrote:

Dear there,

Is it expected to get negative tag count by using DESeq2?

I am using HOMER (getDifferentialPeaksReplicates.pl http://homer.ucsd.edu/homer/ngs/peaksReplicates.html) for my analysis, which uses R/Bioconductor and DESeq2 (by default) to perform the differential enrichment calculations. Why did I get negative tag count for my results? I pasted some of lines from the results:

Inp1_tagDir Tag Count in given bp (14503111.0 Total, normalization factor = 1, effective total = 10000000) Inp2_tagDir Tag Count in given bp (15163786.5 Total, normalization factor = 1, effective total = 10000000) Inp3_tagDir Tag Count in given bp (14010122.5 Total, normalization factor = 1, effective total = 10000000) IP1_tagDir Tag Count in given bp (13945492.5 Total, normalization factor = 1, effective total = 10000000) IP2_tagDir Tag Count in given bp (11841367.0 Total, normalization factor = 1, effective total = 10000000) IP3_tagDir Tag Count in given bp (15415780.0 Total, normalization factor = 1, effective total = 10000000)
-0.288246836848404 0.433387837255659 -0.268014291397154 1.7191399108166 2.70673888043271 1.86558388806041
0.446205758005484 -0.342275173587881 -0.296143075827782 1.39627776477782 2.12637915698719 2.54507683196772
0.695292076927187 -0.0274159575070236 0.72292223174561 2.79295984906655 2.83354767081108 2.21324718854021

 

Does it attribute to "normalization factor"? If using this, is it expected to get negative tag counts (I am running without it now)?

Thanks a lot!

 

ADD COMMENTlink modified 14 months ago • written 14 months ago by xiaofeiwang1982660
0
gravatar for Michael Love
14 months ago by
Michael Love19k
United States
Michael Love19k wrote:

You'll need to show some DESeq2 code. It's not possible to obtain normalized counts from DESeq2 that are negative. You might have better luck asking the HOMER maintainers for help.

ADD COMMENTlink written 14 months ago by Michael Love19k

 

I did send an email and cross my fingers to get response from them.

Here is the command that I used for HOMER, it is very easy and just one-line command.

getDifferentialPeaksReplicates.pl -t IP1_tagDir IP2_tagDir IP3_tagDir -i Inp1_tagDir/ Inp2_tagDir/ bInp3_tagDir/ -genome AGPv3 -gsize 2.2e9 > ../getDiffPeaksRep/Ip_Inp_peaks.txt

We can't get the DESeq2 code from the one-line command. I tried to look the Perl code and tried to print out the deseq2 command, but not successful so far.

I feel so frustrated to get negative tag counts :-(

Thanks a lot!

 

ADD REPLYlink written 14 months ago by xiaofeiwang1982660
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 236 users visited in the last hour