Question

Bioconductor: quantile normalization

0

Entering edit mode

KVC_bioinfo • 0

@kvc_bioinfo-15319

Last seen 6.7 years ago

Hello All,

I have used Bioconductor ‘preprocessCore’ fro quantile normalization. I am trying t understand how does it work with low expressed or zero count genes. Is someone aware of that? Does it remove the low expressed genes?

TIA

normalization quantile preprocesscore • 1.9k views

ADD COMMENT • link updated 6.7 years ago by James W. MacDonald 67k • written 6.7 years ago by KVC_bioinfo • 0

score 2 · Accepted Answer · 2018-03-22

2

Entering edit mode

James W. MacDonald 67k

@james-w-macdonald-5106

Last seen 30 minutes ago

United States

No, it doesn't remove anything. It wouldn't make any sense to do that. If you look at the help page for normalize.quantiles, there is a link to a pdf that explains the procedure.

That said, preprocessCore, and normalize.quantiles are intended for microarray measurements, and using for count data is probably suboptimal.

ADD COMMENT • link 6.7 years ago James W. MacDonald 67k

0

Entering edit mode

Thank you very much. I have used that on RNA-seq counts

ADD REPLY • link 6.7 years ago KVC_bioinfo • 0

0

Entering edit mode

Well, like I said, that is probably not what you want to do. There are established RNA-Seq analysis pipelines, such as DESeq2 and edgeR, which handle the normalization of the count data in a statistically defensible manner. Unless you know what you are doing, it's probably best to use existing tools rather than a hodgepodge of methods borrowed from various analysis pipelines.

ADD REPLY • link 6.7 years ago James W. MacDonald 67k