[Bioc-sig-seq] strandedness bug in Rsamtools?

Steve Lianoglou mailinglist.honeypot at gmail.com
Sat Feb 6 21:57:37 CET 2010


Hi,

On Sat, Feb 6, 2010 at 11:40 AM, Martin Morgan <mtmorgan at fhcrc.org> wrote:
> On 02/06/2010 07:29 AM, Steve Lianoglou wrote:
>> Hi,
>>
>> On Sat, Feb 6, 2010 at 4:17 AM, Steve Lianoglou
>> <mailinglist.honeypot at gmail.com> wrote:
<snip>

>> R> r$strand
>> [1] + +
>> Levels: + - *
>
> I notice that your levels are different from mine; what does
> levels(strand()) say? what's your locale? (sesssionInfo() ;) --
> Rsamtools and strand() in general was trying to be locale-independent,
> but this seems not to be the case...
>
> The workaround is likely to work in a "C" locale, ?Sys.setlocale

Sorry, I pasted the sessionInfo() in the first post and neglected to
do so in the second, but the locale is:

locale:
[1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8

Also:

> ok, I think it's this
>
> ------------------------------------------------------------------------
> r44085 | p.aboyoun | 2010-01-19 15:16:56 -0800 (Tue, 19 Jan 2010) | 2 lines
> Changed paths:
>   M /trunk/madman/Rpacks/BSgenome/R/strand.R
>
> Added support for Rle strand column in strand,DataTable method.
> Also made "+" the first level of strand factor.
> ------------------------------------------------------------------------
>
> which changed the order of strand levels. This will take a bit of
> coordination on our part. Thanks for the reports.

So, being that you're more familiar with these codebases than I could
hope to be, do you think the problem lies strictly in the BSgenome
package? Or should I look at Rsamtools code, too?

Do you reckon running with a checkout of the version of the BSgenome
repo at the revision before r44085 will be OK to sidestep the issue,
or should I backtrack my Rsamtools, too?

I will, of course, try a few different permutations to see what works,
in the meanwhile, so I'm not expecting you to take time to try that
yourself. If you had some intuition, though, I'd appreciate your point
of view :-)

Thanks,
-steve

-- 
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact



More information about the Bioc-sig-sequencing mailing list