multiple testing correction question

[gene_x]

in bioinformatics, you'll probably run into the multiple testing correction problem at some time of the analysis.. I have read several places online, including a nature biotech primer, etc, I can understand the basic knowledge behind it but I still don't know how to use it in my research or real-life work.. my understanding is I can compute P-values and then either Bonferroni adjustment or FDR based q-value or so although i don't know how to do so..

Do you know of a tutorial type documents that can explain in detail with examples of how this is done?

Thanks so much!

[NicoBxl]

In every stat book.
Here an example : http://udel.edu/~mcdonald/statmultcomp.html

[gene_x]

thanks for the reply.. but there is no demo in that link of how to calculate it..

[dariober]

thanks for the reply.. but there is no demo in that link of how to calculate it..

In practice, this can be done in R with p.adjust. Say pvals.txt is a plain text file with a list of pvalues, one per row (that is, a table with one column):

cat pvals.txt 
0.1
0.2
0.3
0.4
0.5
0.6
0.7

In R:

pvals<- read.table('pvals.txt') ## Read file with pvalues
padj<- p.adjust(pvals$V1, method= 'fdr') ## Adjust
write.table(padj, 'pvals.fdr.txt', row.names= FALSE, col.names= FALSE) ## Write to file the corrected pvalues

Hope this helps!
Dario

[gene_x]

thanks, i'll take a look at that in R!