[BioC] distanceToNearest for GRanges?

[Valerie Obenchain]

Just closing the loop here - Janet and I already spoke about this. Given 
the numerous changes made to dependencies in the devel branch porting 
the changes for nearest() and distanceToNearest() to the release branch 
isn't feasible. The next Bioconductor release is in 5 weeks at which 
time the devel branch will become the release.

Valerie


On 08/20/2012 01:48 PM, Valerie Obenchain wrote:
> Hi Janet,
>
> The precede() and follow() code has been completely rewritten in devel 
> (as compared to release). Because nearest() and distanceToNearest() 
> depend on these, updating in release will require the updating of all 
> functions. This might be ok or this might be a problem as other 
> packages (on which the code may depend)  have evolved in the devel 
> branch. I'll look into this and post back when it is either done or I 
> have more information.
>
> Valerie
>
>
>
> On 08/20/2012 11:57 AM, Janet Young wrote:
>> Valerie Obenchain<vobencha at ...>  writes:
>>
>>> Hi Malcom,
>>>
>>> For distanceToNearest() I started with the ranges method in IRanges.
>>> Along the way I did find that distance,GenomicRanges,GenomicRanges was
>>> not handling the situation of '*','-' and '*','+' strand pairs (simply
>>> returning NA). Now those strand combinations are considered compatible
>>> just as '-','-' and '+','+' are.
>>>
>>> Let me know if you run into problems with the method or have other
>>> suggestions.
>>>
>>> Take care,
>>> Valerie
>>>
>>
>>
>> Hi there,
>>
>> I'm revisiting this older string.
>>
>> I'm still using the release versions of GenomicRanges and IRanges (I 
>> want to
>>   spend a little time figuring out the useDevel stuff, but the 
>> pressure is on
>>    to think about biology right now, and it seems it might be better 
>> for me
>>    to stick to the release versions).
>>
>> That means distanceToNearest isn't available for GRanges in the 
>> version I'm
>>   using, so I'm using a variant of Malcolm's function (thanks again!).
>>   However, I realized that I'm running into the same issue with strands
>>   that you brought up, Valerie, that "*"/"+" pairs were not recognized.
>>
>> For now I'll try to remember to "unstrand" my GRanges objects before 
>> using
>> nearest, but I wondered if it's possible to update the nearest 
>> function in
>> the release version of IRanges so that it deals with the strands better?
>>
>> thanks,
>>
>> Janet
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives: 
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: 
> http://news.gmane.org/gmane.science.biology.informatics.conductor