Dear Henrik,
Thank you for your suggestions.
The answer to your first reply is that the security settings prevent
us from using biocLite() etc so we need to download all files and
usually run install scripts.
Following your suggestion I have now installed DNAcopy v 1.19.2 which
caused some problems on the cluster because of a missing
libgfortran.so. (As I have mentioned in an earlier post, I am running
CBS and GLAD on cluster nodes in order to speed up processing.)
I have just started another session and at the moment it seems to
work, the current output is:
Loading required package: DNAcopy
Loading required package: Cairo
Loading required package: GLAD
[1] "Have fun with GLAD"
However, I still need to test the paired analysis with the SNP6.0
arrays, this will take some time.
Here is how the sessionInfo is supposed to look like:
> sessionInfo()
R version 2.8.1 (2008-12-22)
x86_64-unknown-linux-gnu
locale:
C
attached base packages:
[1] tcltk stats graphics grDevices utils datasets
methods
[8] base
other attached packages:
[1] GLAD_1.18.0 Cairo_1.4-5
DNAcopy_1.19.2
[4] biasnp_0.2.43 RODBC_1.2-5
aroma.affymetrix_1.1.0
[7] aroma.apd_0.1.3 R.huge_0.1.7
affxparser_1.14.2
[10] aroma.core_1.1.2 aroma.light_1.13.2
matrixStats_0.1.6
[13] R.rsp_0.4.0 R.httpd_0.1.1
R.filesets_0.5.2
[16] digest_0.3.1 R.cache_0.1.7
R.utils_1.1.6
[19] R.oo_1.4.6 R.methodsS3_1.0.3
Best regards
Christian
On Jun 29, 4:34 pm, Henrik Bengtsson <h...@stat.berkeley.edu> wrote:
> Hi.
>
> On Mon, Jun 29, 2009 at 5:14 AM,
>
>
>
> cstratowa<christian.strat...@vie.boehringer-ingelheim.com> wrote:
>
> > Dear Henrik,
>
> > Meanwhile I have downloaded the latest versions from your repository
> > and installed them on R-2.8.1. Please note that due to our settings it
> > is NOT POSSIBLE for me to use hbInstall("aroma.affymetrix")!!!
>
> > Sadly, now CBS does no longer work at all. I get the message:
>
> > Loading required package: DNAcopy
> > Loading required package: Cairo
> > Error: package 'DNAcopy' does not have a name space
> > In addition: Warning messages:
> > 1: In log(theta/thetaRef, base = 2) : NaNs produced
> > 2: In log(theta * thetaRef, base = 2) : NaNs produced
> > Execution halted
>
> Output from traceback() would be helpful here. I can make a guess
> where/why it occurs, but I'm not sure what you are doing.
>
>
>
>
>
> > Thus it seems that the new versions do not work with R-2.8.1, so I
> > need to downgrade to the old versions.
>
> > Here is the sessionInfo() for the new versions:
> >> sessionInfo()
> > R version 2.8.1 (2008-12-22)
> > x86_64-unknown-linux-gnu
>
> > locale:
> > C
>
> > attached base packages:
> > [1] tcltk stats graphics grDevices utils datasets
> > methods
> > [8] base
>
> > other attached packages:
> > [1] biasnp_0.2.43 RODBC_1.2-5
> > aroma.affymetrix_1.1.0
> > [4] aroma.apd_0.1.3 R.huge_0.1.7
> > affxparser_1.14.2
> > [7] aroma.core_1.1.2 aroma.light_1.13.2
> > matrixStats_0.1.4
> > [10] R.rsp_0.4.0 R.httpd_0.1.1
> > R.filesets_0.5.2
> > [13] digest_0.3.1 R.cache_0.1.7
> > R.utils_1.1.6
> > [16] R.oo_1.4.6 R.methodsS3_1.0.3
>
> Version of DNAcopy?
>
> So, I am quite sure the aroma.* code is working on much older
> versions, but I cannot speak for external packages.
>
> However, in this case aroma.* do assume that DNAcopy has a namespace
> and that was most likely introduced in v1.18.0. You might have to
> upgrade DNAcopy to v1.18.0 or newer. I really recommend v 1.19.2
> because it is now much much faster (really!). For people who can use
> our install scripts, this can be done by fooling bioc to think you
> have a newer R version as follows:
>
> source("http://www.braju.com/R/hbLite.R";);
> biocLite("DNAcopy", rver="2.10.0");
>
> If you cannot use this, you probably have to download the file
> *.tar.gz (*.zip) file and install it manually.
>
> Let us know if this solves the problem.
>
> /Henrik
>
>
>
> > Best regards
> > Christian
>
> > P.S.: I know that this is OSS and everybody is very busy, but
> > nevertheless I would like to mention, that I am not the only person
> > who is not happy with the policy of many open source software systems
> > to supply a new version every 6 months and no longer support older
> > versions. This is no problem when running the software on one's
> > personal PC but often causes problems in company settings where you
> > need to upgrade a lot of systems and the philosophy is not to change a
> > running system. In our case this means to update the servers and the
> > Linux clusters, too, since I have created a package which allows me to
> > run CBS and GLAD on the cluster nodes.
>
> > On Jun 26, 7:35 am, Henrik Bengtsson <h...@stat.berkeley.edu> wrote:
> >> 2009/6/22 cstratowa <christian.strat...@vie.boehringer-ingelheim.com>:
>
> >> > Does this update also work with R-2.8.1?
>
> >> I avoid increasing the dependency in the DESCRIPTION files unless I
> >> know they have to. Currently I think it is roughly R 2.6.x that is
> >> specified. It does not mean it is guaranteed to work. However, I
> >> only support and run redundancy tests on the lastest stable version of
> >> R (now R 2.9.x).
>
> >> Did you try?
>
> >> /Henrik
>
> >> > Best regards
> >> > Christian
>
> >> > On Jun 18, 5:39 pm, Henrik Bengtsson <h...@stat.berkeley.edu> wrote:
> >> >> Hi,
>
> >> >> update aroma.affymetrix et al. and retry.
>
> >> >> /Henrik
>
> >> >> On Thu, Jun 18, 2009 at 5:01 AM,
>
> >> >> cstratowa<christian.strat...@vie.boehringer-ingelheim.com> wrote:
>
> >> >> > Dear all,
>
> >> >> > Below I show you the relevant code fragments of my function "computeCN
> >> >> > ()" to compute copy-number regions using model CBS. This function can
> >> >> > compute copy-number regions for the usual case (reference=NULL), for
> >> >> > pairs of samples (reference="Pairs") or using reference samples
> >> >> > (reference="Reference").
>
> >> >> > I am using this function for a long time for 100K and 500K mapping
> >> >> > arrays w/o any problems. However for the SNP6.0 array this function
> >> >> > works only for the usual case (reference=NULL), but not for the Pairs/
> >> >> > Reference cases.
>
> >> >> > For SNP6.0 array and Pairs/Reference cases I get always the following
> >> >> > error message:
>
> >> >> > Error in this$files[[1]] : subscript out of bounds
> >> >> > Calls: computeCN ... clearCache.AffymetrixCelSet -> getCdf ->
> >> >> > getCdf.AffymetrixCelSet -> getCdf
> >> >> > Execution halted
>
> >> >> > Do you have any idea why I get these errors?
> >> >> > Do I have to change the code for Pairs/Reference cases, and how?
>
> >> >> > Here is the relevant code of function "computeCN()":
>
> >> >> > "computeCN" <- function(object, pheno, arrays=NULL, reference=NULL,
> >> >> > chromosomes=NULL) {
> >> >> > ## get normalized data
> >> >> > cesList <- effectSet(object);
>
> >> >> > ## get optional reference set
> >> >> > if (is.null(reference)) {
> >> >> > refList <- NULL;
> >> >> > } else if (is.character(reference)) {
> >> >> > ## extract reference samples (and data samples if paired)
> >> >> > refList <- list();
> >> >> > for (chiptype in names(cesList)) {
> >> >> > ces <- cesList[[chiptype]];
> >> >> > cesnames <- getNames(ces);
>
> >> >> > ## get row numbers containing reference and data samples
> >> >> > refcol <- which(pheno[,reference] == 1);
> >> >> > datcol <- which(pheno[,reference] == 0);
>
> >> >> > if (reference == "Pairs") {
> >> >> > cesList[[chiptype]] <- extract(ces, datcol);
> >> >> > refList[[chiptype]] <- extract(ces, refcol);
> >> >> > } else if (reference == "Reference") {
> >> >> > cesRef <- extract(ces, refcol);
> >> >> > ceRef <- getAverageFile(cesRef);
>
> >> >> > ## convert single ceRef file into a set of identical files
> >> >> > of same size/length as ces
> >> >> > numarr <- nbrOfArrays(ces);
> >> >> > cesRef <- rep(list(ceRef), numarr); # becomes a list of
> >> >> > CnChipEffectFile:s
> >> >> > cesRef <- newInstance(ces, cesRef, mergeStrands=ces
> >> >> > $mergeStrands, combineAlleles=ces$combineAlleles);
> >> >> > refList[[chiptype]] <- cesRef;
> >> >> > }#if
> >> >> > }#for
> >> >> > }#if
>
> >> >> > ## select model
> >> >> > model <- CbsModel(cesList, refList);
>
> >> >> > ## get index of arrays
> >> >> > if (is.null(arrays)) {
> >> >> > arrays <- indexOfArrays(model);
> >> >> > } else {
> >> >> > arrays <- which(getNames(model)==arrays);
> >> >> > }#if
>
> >> >> > ## fit model with Chromosome Explorer
> >> >> > ce <- ChromosomeExplorer(model, tags=tags);
> >> >> > setParallelSafe(ce, status=TRUE); #necessary for running on cluster
> >> >> > tmp <- process(ce, arrays=arrays, chromosomes=chromosomes,
> >> >> > force=force, verbose=verbose);
>
> >> >> > ## get fitted CN regions from model
> >> >> > cnrs <- getRegions(model, arrays=arrays, chromosomes=chromosomes,
> >> >> > organism="Human", hgVersion="hg18");
>
> >> >> > return(object);
> >> >> > }#computeCN
>
> >> >> > Here is my sessionInfo:
>
> >> >> >> sessionInfo()
> >> >> > R version 2.8.1 (2008-12-22)
> >> >> > x86_64-unknown-linux-gnu
>
> >> >> > locale:
> >> >> > C
>
> >> >> > attached base packages:
> >> >> > [1] tcltk stats graphics grDevices utils datasets
> >> >> > methods
> >> >> > [8] base
>
> >> >> > other attached packages:
> >> >> > [1] biasnp_0.2.41 RODBC_1.2-5
> >> >> > aroma.affymetrix_1.0.1
> >> >> > [4] aroma.apd_0.1.3 R.huge_0.1.6
> >> >> > affxparser_1.14.2
> >> >> > [7] aroma.core_1.0.1 aroma.light_1.11.1
> >> >> > digest_0.3.1
> >> >> > [10] matrixStats_0.1.4 R.rsp_0.4.0
> >> >> > R.httpd_0.1.1
> >> >> > [13] R.cache_0.1.7 R.utils_1.1.3
> >> >> > R.oo_1.4.6
> >> >> > [16] R.methodsS3_1.0.3
>
> >> >> > Thank you in advance for your help.
>
> >> >> > Best regards
> >> >> > Christian
--~--~---------~--~----~------------~-------~--~----~
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.
To post to this group, send email to aroma-affymetrix@googlegroups.com
To unsubscribe from this group, send email to
aroma-affymetrix-unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/aroma-affymetrix?hl=en
-~----------~----~----~----~------~----~------~--~---
