[R] topTable function from LIMMA

[James W. MacDonald]

Hi Roger,

Vallejo, Roger wrote:
> Dear R-Help,
> 
>  
> 
> I am using the function "topTable" from the LIMMA package.  To estimate
> adjusted P-values there are several options (adjust="fdr" , adjust="BH")
> as shown below:
> 
>  
> 
> topTable(fit, number = 10, adjust = "BH", fit$Name)   
> 
>  
> 
> I guess any of these options (fdr, BH, etc.) is using a default of
> FDR=0.05 which is quite conservative (i.e., very likely none of the
> tested genes will be ranked as differentially expressed at 5% FDR). I
> would argue that for an exploratory study it would be okay to use a 0.05
> < FDR <0.20. So, I was wondering if we can change that default of
> FDR=0.0.5 with a less stringent FDR value (0.10, 0.20, etc.) in the
> "topTable" function.  How do I change it?

The topTable() function doesn't have a default cutoff based on the 
p-value (adjusted or otherwise). The cutoff is based on the 'number' 
argument, which has a default of 10, so regardless of the p-value 
adjustment method used, you only get 10 genes if you don't change this 
default.

You can either increase the number to something arbitrarily high and 
then subset the resulting data.frame based on the p-value, or just use 
write.fit() to ouput the whole table and do the work in e.g., Excel.

Best,

Jim


> 
>  
> 
> Thanks a lot for the help on the use of "topTable" function from the
> LIMMA package.
> 
>  
> 
> Roger
> 
>  
> 
>  
> 
> Roger L. Vallejo, Ph.D.
> 
> Computational Biologist & Geneticist
> 
> U.S. Department of Agriculture, ARS          
> 
> National Center for Cool & Cold Water Aquaculture
> 
> 11861 Leetown Road
> 
> Kearneysville, WV 25430
> 
> Voice:    (304) 724-8340 Ext. 2141
> 
> Email:   roger.vallejo at ars.usda.gov <mailto:roger.vallejo at ars.usda.gov> 
> 
>  
> 
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.


-- 
James W. MacDonald, M.S.
Biostatistician
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623


**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.