[BioC] Sorting matrix by column

[James W. MacDonald]

On 10/23/2012 11:15 AM, Guest [guest] wrote:
> Hi,
>
> I would like to sort a matrix by a specific column (column 2). I tried the order() function, but I get an error. I think it is because the values in column 2 are not numeric, they are gene symbols. This may be a general R question, but I thought I would post it here since it is microarray data analysis.
>
> I have matrix x:
>
>> x
>           ID         Gene Symbol     logFC      Adj.PVal
> 10344624 "10371400" "Lypla1"        0.3592492  0.9999522
> 10344633 "10453900" "Tcea1"         0.1886117  0.9999522
> 10344637 "10375051" "Atp6v1h"       0.6713107  0.9999522
> 10344653 "10575211" "Oprk1"         -0.2342731 0.9999522
> 10344658 "10566254" "Rb1cc1"        1.790676   0.9999522
> 10344674 "10602372" "Fam150a"       1.397496   0.9999522
> 10344679 "10398428" "St18"          -0.3278807 0.9999522
> 10344707 "10383518" "Pcmtd1"        -0.2231074 0.9999522
> 10344713 "10397054" "Ahcy"          -0.1844897 0.9999522
> 10344723 "10384020" "Rrs1"          -0.2322781 0.9999522
> 10344725 "10608710" "Adhfe1"        0.5993566  0.9999522
> 10344741 "10363762" "Hnrnpa3"       -0.2660978 0.9999522
> 10344743 "10375058" "3110035E14Rik" 0.9178868  0.9999522
> 10344750 "10381603" "Sgk3"          -0.2961638 0.9999522
> 10344772 "10442373" "6030422M02Rik" -0.1653454 0.9999522
> 10344789 "10421227" "Cspp1"         -0.1480766 0.9999522
> 10344799 "10534966" "Cspp1"         -0.2436361 0.9999522
> 10344801 "10398408" "Cspp1"         -0.4040665 0.9999522
> 10344803 "10398418" "Cspp1"         -0.2556627 0.9999522
> 10344805 "10572772" "Cspp1"         -0.1864641 0.9999522
>
> I want to sort on the "Gene Symbol" column so that I can remove the duplicates and keep the one with the highest log fold change.
>
> I tried the following and received an error.
>> x[order(x[,2]),]
> Error in order(x[, 2]) : unimplemented type 'list' in 'orderVector1'

I am not sure the sessionInfo() you give below corresponds to the 
session above. I get:

 > x <- data.frame(ID = 12345:12354, Gene = 
Rkeys(mogene10sttranscriptclusterSYMBOL)[5001:5010], logFC = rnorm(10), 
pval = runif(10))
 > x
       ID   Gene       logFC      pval
1  12345  Sepw1  0.56914952 0.4916910
2  12346  Serf1  0.83929962 0.4816986
3  12347 Gm4748  0.12462117 0.9372249
4  12348   Sez6 -0.21468480 0.4921201
5  12349  Foxp3 -1.36283694 0.4575675
6  12350  Sfpi1  1.03632565 0.5251826
7  12351  Sfrp1  0.04689108 0.3068112
8  12352   Frzb  0.08379607 0.1509499
9  12353  Sfrp4 -1.61513620 0.9336235
10 12354  Srsf2  1.56222316 0.2571122
 > x[order(x[,2]),]
       ID   Gene       logFC      pval
5  12349  Foxp3 -1.36283694 0.4575675
8  12352   Frzb  0.08379607 0.1509499
3  12347 Gm4748  0.12462117 0.9372249
1  12345  Sepw1  0.56914952 0.4916910
2  12346  Serf1  0.83929962 0.4816986
4  12348   Sez6 -0.21468480 0.4921201
6  12350  Sfpi1  1.03632565 0.5251826
7  12351  Sfrp1  0.04689108 0.3068112
9  12353  Sfrp4 -1.61513620 0.9336235
10 12354  Srsf2  1.56222316 0.2571122

It appears you have something loaded that thinks you want to use the 
orderVector1() function. You can always specify the function you are 
intending with the :: operator (in this case, you want base::order()).

 > x[base::order(x[,2]),]
       ID   Gene       logFC      pval
5  12349  Foxp3 -1.36283694 0.4575675
8  12352   Frzb  0.08379607 0.1509499
3  12347 Gm4748  0.12462117 0.9372249
1  12345  Sepw1  0.56914952 0.4916910
2  12346  Serf1  0.83929962 0.4816986
4  12348   Sez6 -0.21468480 0.4921201
6  12350  Sfpi1  1.03632565 0.5251826
7  12351  Sfrp1  0.04689108 0.3068112
9  12353  Sfrp4 -1.61513620 0.9336235
10 12354  Srsf2  1.56222316 0.2571122

Best,

Jim


>
> If anyone has any suggestions for an easy way to sort a significant gene list, remove duplicated values, and keep the value with highest fold change, that would be helpful!
>
> I've posted my session info below.
>
> Thanks!
>
> Guest
>
>   -- output of sessionInfo():
>
>> sessionInfo()
> R version 2.15.1 (2012-06-22)
> Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
>
> locale:
> [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> loaded via a namespace (and not attached):
> [1] tools_2.15.1
>
> --
> Sent via the guest posting facility at bioconductor.org.
>
> _______________________________________________
> 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