GWASTools assocTestCPH on cluster vs. computer
0
Entering edit mode
mrynerson • 0
@mrynerson-6882
Last seen 6.3 years ago
United States

Hello,

I am getting the following error after setting up a GenotypeData object in GWASTools:

Error in assocTestCPH(genoData, event = "pheno", time.to.event = "time_to_event",  : 

  required phenotypes are not in genoData

I had to add the phenotypes to the scanAnnot object manually from a data frame, as they were not read in by imputedDosageFile(), despite being in the .sample file (ncdfImputedDosage worked for this, but is deprecated). 

However, they are in genoData: when I type getScanVariableNames(genoData), I see my phenotypes, and they are listed under

@scannAnnot

    @data

           $pheno

in str(genoData). In addition, the exact same script works perfectly well on my laptop - no error. I am trying to get this to run on a cluster so I can run the analysis over an array of chromosome fragments.

Does anyone have an idea of why this would run differently in an R session on a cluster vs. my computer?

Thanks!

 

sessionInfo() for my computer:

R version 3.1.1 (2014-07-10)
Platform: x86_64-apple-darwin13.1.0 (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] splines   parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] snpStats_1.16.0     Matrix_1.1-4        survival_2.37-7     GWASTools_1.12.0    ncdf_1.6.8          Biobase_2.26.0     
[7] BiocGenerics_0.12.1 gdsfmt_1.1.0.1     

loaded via a namespace (and not attached):
 [1] DBI_0.3.1          DNAcopy_1.40.0     grid_3.1.1         GWASExactHW_1.01   lattice_0.20-29    lmtest_0.9-33     
 [7] quantreg_5.05      quantsmooth_1.32.0 RSQLite_1.0.0      sandwich_2.3-2     SparseM_1.05       tools_3.1.1       
[13] zlibbioc_1.12.0    zoo_1.7-11        

 

sessionInfo() for my cluster session:

R version 3.1.1 (2014-07-10)

Platform: x86_64-unknown-linux-gnu (64-bit)

 

locale:

 [1] LC_CTYPE=en_US       LC_NUMERIC=C         LC_TIME=en_US       

 [4] LC_COLLATE=en_US     LC_MONETARY=en_US    LC_MESSAGES=en_US   

 [7] LC_PAPER=en_US       LC_NAME=C            LC_ADDRESS=C        

[10] LC_TELEPHONE=C       LC_MEASUREMENT=en_US LC_IDENTIFICATION=C 

 

attached base packages:

 [1] splines   parallel  grid      stats     graphics  grDevices utils    

 [8] datasets  methods   base     

 

other attached packages:

 [1] snpStats_1.16.0     Matrix_1.1-4        survival_2.37-7    

 [4] GWASTools_1.12.0    Biobase_2.26.0      BiocGenerics_0.12.0

 [7] quantsmooth_1.32.0  quantreg_5.05       SparseM_1.05       

[10] DNAcopy_1.40.0      GWASExactHW_1.01    gdsfmt_1.1.2       

[13] ncdf_1.6.8         

 

loaded via a namespace (and not attached):

[1] DBI_0.3.1       lattice_0.20-29 lmtest_0.9-33   RSQLite_1.0.0  

[5] sandwich_2.3-2  zlibbioc_1.12.0 zoo_1.7-11     

gwastools • 903 views
ADD COMMENTlink
0
Entering edit mode

Thanks for reporting the bug about data from the .sample file not appearing in the scan annotation.  I will work on a fix for this.

I am perplexed about the different behavior between your Mac and your cluster.  Can you try the following on the cluster:

cvnames <- unique(unlist(strsplit(covars,"[*:]")))
c(event,time.to.event,cvnames)
getScanVariableNames(genoData)
is.element(c(event,time.to.event,cvnames), getScanVariableNames(genoData))

and paste the results?  The test in the last line is what triggers the error you reported, but I don't know why it would behave differently on different machines.

ADD REPLYlink
0
Entering edit mode

Thank you Stephanie. Somehow it is working today, I don't know why! 

The phenotype problem is still there though, imputedDosageFile only returns scanID, sampleID, and added.

ADD REPLYlink
0
Entering edit mode
@stephanie-m-gogarten-5121
Last seen 11 weeks ago
University of Washington

The bug in reading the samples file for IMPUTE2 is fixed in release 1.12.1 and devel 1.13.3.

ADD COMMENTlink
0
Entering edit mode
eakhram • 0
@eakhram-7271
Last seen 6.1 years ago
United States

Dear all,
I would like to run assocTestCPH with imputed data using IMPUTE2 and drug dose to event (as phenotype), but I am not familiar with this function. Could someone, please share an example of how to construct GenotypeData file? 

Thank you, 
Katya  

ADD COMMENTlink
0
Entering edit mode
@stephanie-m-gogarten-5121
Last seen 11 weeks ago
University of Washington

The function imputedDosageFile will convert IMPUTE2 files into the format needed for GWASTools (a GDS file and SNP and scan annotation objects, which are combined into a GenotypeData object). Look at the help page for that function for an example. You could also look at the vignette "Data formats in GWASTools" for more information.

ADD COMMENTlink

Login before adding your answer.

Similar Posts
Loading Similar Posts
Traffic: 142 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.4