[BioC] 'Rsamtools' produces segfaults with XS tag

[Martin Morgan]

Thanks Misha for the additional hints; I looked at the code and couldn't
see anything immediately wrong, but if you can provide a (minimal?)  bam
file displaying the problem that would be helpful. Martin

Misha Kapushesky <ostolop at ebi.ac.uk> writes:

> Hi,
>
> I can add more detail to this - the same problem:
>
>> library( "Rsamtools" )
> ...
>> tags <- c("NM", "MD", "XA", "XS","NS")
>> and_file = "SRR034792/tophat_out/accepted_hits.sortednames.bam"
>> and_bam = scanBam(and_file, param = ScanBamParam(flag = scanBamFlag(isUnmappedQuery = FALSE), tag = tags))
>
> *** caught segfault ***
> address 0x2b3449beb8cd, cause 'invalid permissions'
>
> Traceback:
> 1: .Call(func, file, index, "rb", NULL, flag, simpleCigar, ...)
> 2: .io_bam(.scan_bam, file, index, reverseComplement, tmpl, param = param)
> 3: scanBam(and_file, param = ScanBamParam(flag = scanBamFlag(isUnmappedQuery = FALSE),     tag = tags))
> 4: scanBam(and_file, param = ScanBamParam(flag = scanBamFlag(isUnmappedQuery = FALSE),     tag = tags))
>
> This error goes away if we don't include "XS" in tags, also if we use
> version 1.0.0 of Rsamtools. Any ideas?
>
> The sessionInfo() output is below. Thanks in advance!
>
> --Misha
>
>> sessionInfo()
> R version 2.11.1 (2010-05-31)
> x86_64-unknown-linux-gnu
>
> locale:
> [1] C
>
> attached base packages:
> [1] tools     stats     graphics  grDevices utils     datasets  methods
> [8] base
>
> other attached packages:
> [1] Rsamtools_1.1.11    Biostrings_2.16.9   GenomicRanges_1.0.7
> [4] IRanges_1.6.11      svMisc_0.9-57       JavaGD_0.5-2
> [7] rJava_0.8-4
>
> loaded via a namespace (and not attached):
> [1] Biobase_2.8.0
>
>
> On Thu, 12 Aug 2010, Andrew Tikhonov wrote:
>
>> Dear Biocore Team,
>>
>> 	We've got a problem where 'Rsamtools' produces segfaults and
>> core-dumps when reading a .bam file using the "XS" tag. We have a
>> faulty .bam and the list or R commands to reproduce it. Could you
>> please point us to someone who we may discuss this matter with and,
>> if possible, fix the problem.
>>
>> 	Thanks a lot in advance.
>>
>> Regards,
>> Andrew
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at stat.math.ethz.ch
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

-- 
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109

Location: Arnold Building M1 B861
Phone: (206) 667-2793