[BioC] Filter before? Filter after? With maanova

[Loren Engrav]

I was afraid you would say that, ie must filter before fitting

I have been filtering on human expression "by hand" and more probe sets to
examine is lots of work; so letting Fs cut first saves time; but if not
appropriate then not appropriate

So yesterday I set up a Filemaker script that passed testing and so hit the
"do" button; is now 50% done after 1.5 hours but this will be better than
the manual method

So will filter before fitting

Thank you


> From: Robert Gentleman <rgentlem at fhcrc.org>
> Date: Mon, 6 Jul 2009 09:43:32 -0700
> To: Loren Engrav <engrav at u.washington.edu>
> Cc: "bioconductor at stat.math.ethz.ch" <bioconductor at stat.math.ethz.ch>
> Subject: Re: [BioC] Filter before? Filter after? With maanova
> 
> Hi Loren,
>   The only results that I am aware of suggest that using non-specific
> filtering (as described in the Case Studies book, and a hopefully soon
> to appear paper) is in general of benefit.
>    Filtering after you fit a model is not appropriate - at that point
> you must rely on what ever form of p-value adjustment you prefer.
> 
>    best wishes
>      Robert
> 
> 
> On Sat, Jul 4, 2009 at 5:07 PM, Loren Engrav<engrav at u.washington.edu> wrote:
>> I have 10,000 probe sets that I am processing with mixed linear regression
>> and the maanova package
>> I can filter in various ways, ie does human expression match porcine
>> expression?, and reduce the number of probe sets
>> Is it more valid to filter before doing the regression, or after?
>> Thank you
>> --
>> Loren Engrav, MD
>> Univ Washington
>> Seattle
>> 
>> 
>> 
>>        [[alternative HTML version deleted]]
>> 
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>> 
>> 
> 
> 
> 
> -- 
> Robert Gentleman, PhD
> Program in Computational Biology
> Division of Public Health Sciences
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N, M2-B876
> PO Box 19024
> Seattle, Washington 98109-1024
> 206-667-7700
> rgentlem at fhcrc.org