[BioC] mapping multi-mapping reads in Rsubread (summarizeOverlaps mode ignoring inter feature overlaps)

[Wei Shi]

Dear Tom,

Sorry, it took longer than I originally thought to add an option to Rsubread to make it output multiple mapping locations for multi-mapping reads. It has just been implemented in Rsubread 1.11.1 in Bioc devel branch. The 'nBestLocations' parameter in align function now allows you to report up to 16 equally-best locations for those multi-mapping locations. It will take a couple of days before you can access this version. Alternatively, you can use the C version of Subread aligner (http://subread.sourceforge.net)  straight away to get multiple mapping locations for multi-mapping reads.

We have tested the new versions of these programs and they worked fine.

Best regards,
Wei


On Apr 11, 2013, at 3:35 PM, Thomas Girke wrote:

> Excellent. And thanks for doing this.
> 
> Thomas
> 
> On Thu, Apr 11, 2013 at 03:16:44AM +0000, Wei Shi wrote:
>> Dear Tom,
>> 
>> Please see my reply below.
>> 
>> On Apr 11, 2013, at 12:55 PM, Thomas Girke wrote:
>> 
>>> Dear Wei,
>>> 
>>> Thanks. I guess this makes sense, but is it possible this way to perform
>>> transcript level counting in addition to exonic gene-level counting? 
>>> 
>> 
>> Yes, it can do that. But you will have to provide your own transcript annotation. Rsubread only includes RefSeq annotation for exons.
>> 
>>> Also, this reminds me of one question related to Rsubread for which I
>>> couldn't find an answer in the documentation (haven't had time to read
>>> your NAR paper though). Is it possible to use Rsubread for aligning
>>> directly against transcriptome sequences where we want to set no
>>> restrictions on multiple mappings of equal quality in order to obtain
>>> transcript read counts? Right I am often doing this with Bowtie2 for our
>>> own transcriptome assemblies where we don't have a reference genome and
>>> then use Rsamtools for counting the reads mapping to individual
>>> transcripts. With Rsubread I was not sure if the algorithm can return
>>> any number of alternative hits? If it can then I would love to give it 
>>> a try for our transcriptome assembly projects.
>>> 
>>> Thomas
>>> 
>>> 
>> 
>> Rsubread reports no more than one mapping location for each read. But we probably should provide an option to allow multiple mapping locations to be reported for the reads. This should be pretty straightforward to do. I will let you know once we added this option (probably in a couple of days).
>> 
>> Cheers,
>> Wei
>> 
>> 
>> 
>> 
>> 
>> 
>> ______________________________________________________________________
>> The information in this email is confidential and intended solely for the addressee.
>> You must not disclose, forward, print or use it without the permission of the sender.
>> ______________________________________________________________________

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:6}}