CalMaTe model fitting error

[thomas.g...@merckgroup.com]

Hello,

I am using CALMATE 0.12.1 to improve ASCN for a 1213 sample SNP6 data set without reference samples.

I run the analysis on a 32core 156Gb main memory server. When CALMATE is fitting the model the following error occurs:

...

setOption(aromaSettings, "memory/ram", 10.0)

...

cmt <- CalMaTeCalibration(dsNList)

dsCList <- process(cmt, verbose=verbose)

...

20180913 09:41:43|     Fitting CalMaTe...

20180913 09:41:43|      Algorithm flavor: v2

20180913 09:41:43|      Number of SNPs: 515251

20180913 09:41:43|      Number of SNPs left: 515251, 514751, 514251, 513751, 513251, 512751, 512251, 511751, 511251, 510751, 510251, 509751, 509251, 508751, 508251, 507751, 507251, 506751, 506251, 505751, 505251, 504751, 504251, 503751, 503251, 502751, 502251, 501751, 501251, 500751, 500251, 499751, 49925

...

288251, 287751, 287251, 286751, 286251, 285751, 285251, 284751, 284251, 283751, 283251, 282751, 282251, 281751, 281251, 280751, 280251, 279751, 279251, 278751, 278251, 277751, 277251, Error in qr.solve(t(TR) * wgths^2, t(Genotypes) * wgths^2) : 

  singular matrix 'a' in solve

I tried a second run which leads to a similar error:

...

20180913 15:49:54|      All data points are finite.

20180913 15:49:54|     Identifying non-finite data points...done

20180913 15:49:54|     Fitting CalMaTe...

20180913 15:49:54|      Algorithm flavor: v2

20180913 15:49:54|      Number of SNPs: 51525

20180913 15:49:54|      Number of SNPs left: 51525, 51025, 50525, 50025, 49525, 49025, 48525, 48025, 47525, 47025, 46525, 46025, 45525, 45025, 44525, 44025, 43525, 43025, 42525, 42025, 41525, 41025, 40525, 40025, 39525, 39025, 38525, 38025, 37525, 37025, 36525, 36025, 35525, 35025, 34525, 34025, 33525, 33025, 32525, 32025, 31525, 31025, 30525, 30025, 29525, 29025, 28525, 28025, 27525, 27025, 26525, 26025, 25525, 25025, 24525, 24025, 23525, 23025, 22525, 22025, 21525, 21025, 20525, 20025, 19525, Error in qr.solve(t(TR) * wgths^2, t(Genotypes) * wgths^2) : 

  singular matrix 'a' in solve

And I get then the downstream error when I try to access the fitted fractionB data:

dsCTxt<- writeDataFrame(dsCList$fracB, columns=c("unitName", "chromosome", "position", "*"))

Error in writeDataFrame(dsCList$fracB, columns = c("unitName", "chromosome",  :  object 'dsCList' not found

Any suggestion how to find the cause of error and a resolution ?

> sessionInfo()

R version 3.5.1 (2018-07-02)

Platform: x86_64-w64-mingw32/x64 (64-bit)

Running under: Windows 7 x64 (build 7601) Service Pack 1

Matrix products: default

locale:

[1] LC_COLLATE=German_Germany.1252  LC_CTYPE=German_Germany.1252    LC_MONETARY=German_Germany.1252 LC_NUMERIC=C                   

[5] LC_TIME=German_Germany.1252    

attached base packages:

[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:

 [1] calmate_0.12.1         aroma.light_3.10.0     aroma.affymetrix_3.1.1 aroma.core_3.1.3       R.devices_2.16.0      

 [6] R.filesets_2.12.1      R.utils_2.7.0          R.oo_1.22.0            affxparser_1.52.0      R.methodsS3_1.7.1     

loaded via a namespace (and not attached):

 [1] matrixStats_0.54.0 codetools_0.2-15   listenv_0.7.0      future_1.9.0       digest_0.6.16      MASS_7.3-50        R.huge_0.9.0      

 [8] future.apply_1.0.1 PSCBS_0.64.0       tools_3.5.1        R.cache_0.13.0     parallel_3.5.1     compiler_3.5.1     base64enc_0.1-3   

[15] aroma.apd_0.6.0    R.rsp_0.43.0       globals_0.12.2     DNAcopy_1.54.0    

best regards

Thomas

[thomas.g...@merckgroup.com]

I forgot the traceback:

traceback()

11: stop("singular matrix 'a' in solve")

10: qr.solve(t(TR) * wgths^2, t(Genotypes) * wgths^2)

9: fitCalMaTeMedians(dataInit, references = references, fB1 = fB1, 

       fB2 = fB2)

8: fitFcn(Cjj, references = references, ...)

7: calmateByThetaAB.array(theta, references = references, ..., verbose = verbose)

6: calmateByThetaAB(theta, references = references, ..., verbose = verbose)

[Henrik Bengtsson]

Hi. This is a hard-to-troubleshoot problem. This could be due to one
or more arrays having weak/noisy signals, or possibly even being
corrupt.

As a first small step, I've created feature request
https://github.com/HenrikBengtsson/calmate/issues/7. When that is in
place, one can narrow down the problematic data and make the estimator
more robust.

/Henrik

> --
> --
> When reporting problems on aroma.affymetrix, make sure 1) to run the latest version of the package, 2) to report the output of sessionInfo() and traceback(), and 3) to post a complete code example.
>
>
> You received this message because you are subscribed to the Google Groups "aroma.affymetrix" group with website http://www.aroma-project.org/.
> To post to this group, send email to aroma-af...@googlegroups.com
> To unsubscribe and other options, go to http://www.aroma-project.org/forum/
>
> ---
> You received this message because you are subscribed to the Google Groups "aroma.affymetrix" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to aroma-affymetr...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.