search

HCBravoLab / metagenomeSeq

Statistical analysis for sparse high-throughput sequencing
66 stars 20 forks source link

Different results with cumNormMat and MRcounts(norm=TRUE) #69

Closed alisimecn closed 5 years ago

alisimecn commented 5 years ago

Hi !

I'm having an issue when trying to normalize my counts : I get a different matrix when I use cumNormMat(mydata, p=p) and when I use MRcounts(mydata, norm=TRUE, log=TRUE).

Here, mydata is an MRexperiment object obtained from a biom file.

Weirdly enough, I get the same matrix by doing MRcounts(mydata, norm=TRUE, log=TRUE) and MRcounts(cumNorm(mydata), norm=TRUE, log=TRUE).

Did I misunderstand the functions cumNormMat and MRcounts ? Are they not supposed to do the same thing ?

Thank you so much for your help,

Cheers !

jnpaulson commented 5 years ago

cumNormMat doesn't report log transformed counts, while MRcounts does. You can log transform following cumNormMat

On Thu, Apr 11, 2019 at 5:44 AM alisimecn notifications@github.com wrote:

Hi !

I'm having an issue when trying to normalize my counts : I get a different matrix when I use cumNormMat(mydata, p=p) and when I use MRcounts(mydata, norm=TRUE, log=TRUE).

Here, mydata is an MRexperiment object obtained from a biom file.

Weirdly enough, I get the same matrix by doing MRcounts(mydata, norm=TRUE, log=TRUE) and MRcounts(cumNorm(mydata), norm=TRUE, log=TRUE).

Did I misunderstand the functions cumNormMat and MRcounts ? Are they not supposed to do the same thing ?

Thank you so much for your help,

Cheers !

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/HCBravoLab/metagenomeSeq/issues/69, or mute the thread https://github.com/notifications/unsubscribe-auth/AC5hg8BhHVuVR83jq28iVwXwS22cxTEQks5vfy42gaJpZM4cphL6 .

alisimecn commented 5 years ago

Oh, okay, thank you very much!