[BioC] htSeqTools errors

[Juan L. Mateo]

Hi Oscar,

I am aware of the purpose of the cmds function but I got the same error 
when I have several samples, so, for simplicity, I posted this example 
with only one sample.

I am already using the development version as you can see from the 
sessionInfo() output at the end of my first email.

As I said, I can send you the R objects or bam files, just let me know 
in which way I should do that since they are too big to be attached in 
an email

Best,
Juan

On 02/21/2012 10:33 AM, Oscar Reina wrote:
> Hi Juan,
>
> we have been working on the giniCoverage function regarding several
> issues that arised with version 1.0 of the package. They are already
> available in the development version of the package available from the
> Bioconductor development repository. Please let me know should you have
> doubts or experience problems installing that version.
>
> As for the cmds function, from your code I assume your data is basically
> a single sequencing sample, whereas the idea behind the cmds function is
> to produce a pca-like plot showing separation and relative positioning
> of several samples to assess their similarities. Nonetheless I'm not
> sure if the error you are getting is due to that, and would be grateful
> if you could give me access to your .bam file is possible so I can try
> to reproduce the error here.
>
> Best regards,
>
> Oscar
>
> On Tue, Feb 21, 2012 at 10:12 AM, Juan L. Mateo
> <juan.mateo at cos.uni-heidelberg.de
> <mailto:juan.mateo at cos.uni-heidelberg.de>> wrote:
>
>     Hi Valerie,
>
>     Thanks for you answer.
>     This object contains 16M reads and I am happy to provide any
>     information that would be useful or even this object which is 206Mb big.
>
>      > sapply(samples,nrow)
>     Con_input
>       16513449
>
>     Best,
>     Juan
>
>
>     On 02/20/2012 08:08 PM, Valerie Obenchain wrote:
>
>         Hi Juan,
>
>         ccing the author in case he hasn't seen this.
>
>         Can you provide more information on what 'samples' looks like or the
>         object itself?
>
>         Valerie
>
>
>
>
>         On 02/16/12 04:51, Juan L. Mateo wrote:
>
>             Hi all,
>
>             I am trying to use htSeqTools to analyse some ChIP-seq
>             samples and I
>             am getting errors on the methods cmds and giniCoverage.
>             Below I paste
>             the output of a session.
>
>             Any help will be greatly appreciated.
>
>
>              > what=c("rname", "strand", "pos", "qwidth")
>              > param=ScanBamParam(what = what)
>              >
>              > Con_input<-scanBam('Con_input.__fastq.bam', param=param)[[1]]
>              >
>             Con_input.reads<-GRanges(__seqnames=Rle(Con_input$rname),__ranges=IRanges(Con_input$pos,
>             width=Con_input$qwidth))
>              >
>             samples<-RangedDataList(Con___input=as(Con_input.reads,"__RangedData"))
>              > cmds1<-cmds(samples,k=2)
>             Computing coverage...
>             Computing correlations...
>             Error in d[[i]] : subscript out of bounds
>              > ssdCoverage(samples)
>             Con_input
>             0.2551517
>              > giniCoverage(samples,mc.cores=__1)
>             Simulating uniformily distributed data
>             Error in sample.int <http://sample.int>(chrLen[i],
>             as.integer(chrReads[i]), replace = T) :
>             invalid 'size' argument
>             In addition: Warning messages:
>             1: In FUN(X[[1L]], ...) : no non-missing arguments to max;
>             returning -Inf
>             2: In FUN(X[[1L]], ...) : no non-missing arguments to max;
>             returning -Inf
>             3: In FUN(X[[1L]], ...) : no non-missing arguments to max;
>             returning -Inf
>             4: In eval(expr, envir, enclos) : NAs introduced by coercion
>
>
>
>
>              > sessionInfo()
>             R Under development (unstable) (2012-02-13 r58337)
>             Platform: x86_64-unknown-linux-gnu (64-bit)
>
>             locale:
>             [1] LC_CTYPE=en_GB LC_NUMERIC=C LC_TIME=en_GB
>             [4] LC_COLLATE=en_GB LC_MONETARY=en_GB LC_MESSAGES=en_GB
>             [7] LC_PAPER=C LC_NAME=C LC_ADDRESS=C
>             [10] LC_TELEPHONE=C LC_MEASUREMENT=en_GB LC_IDENTIFICATION=C
>
>             attached base packages:
>             [1] stats graphics grDevices utils datasets methods base
>
>             other attached packages:
>             [1] htSeqTools_1.1.2 BSgenome_1.23.2 MASS_7.3-17
>             [4] Biobase_2.15.3 Rsamtools_1.7.33 Biostrings_2.23.6
>             [7] GenomicRanges_1.7.24 IRanges_1.13.24 BiocGenerics_0.1.4
>
>             loaded via a namespace (and not attached):
>             [1] bitops_1.0-4.1 zlibbioc_1.1.1
>
>
>
>
>
>
> --
> Oscar Reina
> Research officer, Bioinformatics and Biostatistics unit
> IRB Barcelona
> Cluster building, PBC32
> Tel (+34) 93 402 0553
> Fax (+34) 93 402 0257
>
> oscar.reina at irbbarcelona.org <mailto:oscar.reina at irbbarcelona.org>
> http://www.irbbarcelona.org/bioinformatics