Search
Question: Filtering array data with nsFilter
0
gravatar for emarram1
2.2 years ago by
emarram10
emarram10 wrote:

I am filtering the data that I am working with with nsFilter, 

filtered <- nsFilter(eset_rma, require.entrez = TRUE, remove.dupEntrez = TRUE, var.func = IQR ,var.cutoff = 0.5,var.filter = TRUE,filterByQuantile = FALSE, feature.exclude = "^AFFX")

 

when I use filterByQuantile = FALSE, instead of TRUE, what is this doing to my data?

after filtering I am using the limma package to decide which genes are different in each condition, and when I use filterByQuantile = FALSE I get more genes, why is that?

 

ADD COMMENTlink modified 2.2 years ago by Gordon Smyth34k • written 2.2 years ago by emarram10
2
gravatar for Gordon Smyth
2.2 years ago by
Gordon Smyth34k
Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia
Gordon Smyth34k wrote:

Please don't do this filtering. Using nsFilter() with var.filter=TRUE will sabotage any DE analysis you might do using limma. It would be better to take a simpler common-sense filtering approach, or even not to filter at all.

Have a look at previous posts on this topic, for example:

  Is filtering necessary to extract significant DE genes with limma ?

ADD COMMENTlink modified 2.2 years ago • written 2.2 years ago by Gordon Smyth34k
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: 193 users visited in the last hour