[miso-users] 0 reads align to all events

[Yarden Katz]

Hi Konstantinos,

Thanks for this note -- it's true that the BAM file must be sorted and indexed.  I have never used Galaxy so I'm not sure if that's the issue.

Another potential cause: samtools/pysam which is used by MISO to access the BAM is very particular about the index and its BAM being associated in the same way in the directory and having identical lower case extensions.  If you have:

mydir/mybam.bam
mydir/mybam.BAM.BAI

then I don't believe it will associate "mybam.BAM.BAI" with "mybam.bam" -- it needs to be "mybam.bam.bai".  Note lower case difference.  I'll add checks for this.

The ultimate way to check if MISO can access your reads is to take an event (e.g. coordinates from SE) that you know should have reads in it, and try:

samtools view mybam.bai chr:start-end

where chr:start-end are the coordinates of the event.  If you cannot access the reads this way, then MISO won't be able to find them either.

Best, --Yarden

On Jan 15, 2013, at 10:17 AM, Charizanis,Konstantinos wrote:

> Good morning. 
> 
> I had the same issue and I was wasting my time trying to figure out what was wrong with my headers. I used every header that was available out there but still nothing. The problem was that I was downloading the BAM file from the galaxy server but the problem was that I was downloading just the normal bam file. Our galaxy here at University of Florida is a bit tricky and does not give you the address of the sorted BAM file that you need. I ended up using the SAM files and converted them to BAM in command line with the provided samtools. That was giving me three files including the necessary sorted.bam file. Maybe this is your problem.
> 
> Just a thought.
> 
> Konstantinos charizanis
> 
> -----Original Message-----
> From: miso-users-bounces at mit.edu [mailto:miso-users-bounces at mit.edu] On Behalf Of Yarden Katz
> Sent: Tuesday, January 15, 2013 10:12 AM
> To: Jian Duke
> Cc: miso-users at mit.edu
> Subject: Re: [miso-users] 0 reads align to all events
> 
> Hi Jian,
> 
> I use MISO on Tophat output regularly so it should not require pre-processing.  Are you sure that your Tophat BAMs match the headers (i.e. chromosomes names) of SE.mm9.gff?  I.e. if you do:
> 
> samtools view yourbam.bam | head
> 
> Do the alignments align to a "chr*" style chromosome?  
> 
> Also, is your --read-len argument correct?  MISO needs to be given the right read length argument.  If your reads are 50bp and you feed it a different number in --read-len, it will not work.
> 
> --Yarden
> 
> 
> On Jan 14, 2013, at 9:57 PM, Jian Duke wrote:
> 
>> Hi Yarden,
>> I ran MISO with the Tophat output generated by the on-line server Galaxy. But I got 0 reads align to all events. I have read the FAQ of MISO manual, but I still do not know what is wrong.
>> What I did include 1) put the downloaded Tophat output files, both .BAM and the .BAI files, in the same folder. 2) I indexed SE.mm9.gff (I only carried out analysis for SE events). Should I process the output files of Tophat for MISO?
>> Thanks.
>> Jian 
>> _______________________________________________
>> miso-users mailing list
>> miso-users at mit.edu
>> http://mailman.mit.edu/mailman/listinfo/miso-users
> 
> 
> _______________________________________________
> miso-users mailing list
> miso-users at mit.edu
> http://mailman.mit.edu/mailman/listinfo/miso-users