[BioC] extracting transcripts

[James W. MacDonald]

Hi Jahn,

On Tuesday, January 14, 2014 9:17:56 AM, Jahn Davik wrote:
> Hi,
> In edgeR I do something like this:
>
> lrt<-glmLRT(fit,contrast=c(0,0,0,0,0,0,0,0,-1,1))
> topTags(lrt)
>
> Coefficient:  -1*GroupE:Time96 1*GroupJ:Time96
>                       logFC logCPM     LR    PValue       FDR
> comp369941_c0_seq1   17.19   9.98 1103.8 4.85e-242 3.05e-237
> comp436540_c0_seq6   13.42   3.92  216.0  6.91e-49  2.17e-44
> comp438623_c0_seq2  -10.88   5.34  199.7  2.43e-45  5.08e-41
> comp437945_c0_seq4  -11.37   4.27  184.9  4.06e-42  6.38e-38
> comp437052_c0_seq13 -10.57   6.99  147.1  7.64e-34  9.60e-30
> comp434611_c0_seq48 -11.08   3.61  116.7  3.34e-27  3.50e-23
> comp423571_c0_seq13 -10.68   2.95  114.3  1.10e-26  9.90e-23
> comp419207_c0_seq15  -2.94   6.34  113.6  1.63e-26  1.28e-22
> comp430114_c0_seq11  -3.42   6.59   97.3  5.82e-23  3.97e-19
> comp432078_c0_seq11 -11.19   4.11   97.2  6.31e-23  3.97e-19
>
>
> But I would like to extract the, say, 50 top ids or the ids that have an FDR below a certain limit. How do I do that?

Do you want the IDs or the table?

Anyway, this is just a subsetting operation that you should be familiar 
with if you are planning to use R for much of anything.

tab <- topTags(lrt, Inf)
tab <- tab[tab$FDR < 0.05]
tab[1:50,]
ids <- row.names(tab)[tab$FDR < 0.05]

etc, etc.

Best,

Jim


>
> Best
> Jahn
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

--
James W. MacDonald, M.S.
Biostatistician
University of Washington
Environmental and Occupational Health Sciences
4225 Roosevelt Way NE, # 100
Seattle WA 98105-6099