BAM file from the 2.6 version

[litali]

Hi
In the 454 2.6 mapper version there is a .bam file created. How can it be viewed in ucsc or in other softwares?

[maubp]

Like any other BAM file? BAM is the binary version of SAM, see

http://samtools.sourceforge.net/

http://samtools.sourceforge.net/swlist.shtml

[litali]

BAM file

I tried different softwares, including aremis bamview, but when i upload the file I just don't see there anything...

[maubp]

Have you indexed the BAM file with "samtools index example.bam"?

What does "samtools idxstats" tell you? This should summarise how many reads were mapped - perhaps very few mapped.

Also the first few lines in SAM format would be interesting for diagnosis, try "samtools view example.bam | head -n 50" for the first 50 lines. If you post them here, wrap it with [ code ] and [ /code ] tags to get it to display nicely on the forum.

[ulz_peter]

BAM files from the GsMapper software is neither reference-ordered nor indexed, so you need to do these steps before actually being able to use it properly.

You can download picard (http://sourceforge.net/projects/picard/) and invoke the SortSam.jar Tool like

java -jar SortSam.jar SO=coordinate INPUT=454Contigs.bam OUTPUT=output.bam VALIDATION_STRINGENCY=LENIENT CREATE_INDEX=true

That should work and most software should be able to visualize it properly. For efficient visualization I recommend IGV (http://www.broadinstitute.org/software/igv/home)

Hope that helps

[litali]

Thank you,
I tried the picard but received many errors in the following format:
Ignoring SAM validation error: ERROR: Read name F01BJ5E01DP1XH, No M or N operator between pair of I operators in CIGAR
????

[ulz_peter]

I get the same error, and by visualizing the 454 Bam files it seems that instead of a substitution it declares variants as an insert and deletion, which makes absolutely no sense. But I'm afraid you can't replace the cigar string without extensive programming...

So until this has been solved you can either align your reads with different software or live with that problem...

[litali]

bam

so, I don't understand, how do you work with this? I should ignore the errors and take the output files (which ones?) to the viewers?

[litali]

P.s I only see one output file there: output.bai

[ulz_peter]

I actually stopped using the BAM file for visualizazion as (at least in IGV) the insertion/deletions are a mess to look at. I now do analysis on the gsMapper without visualization as well as analysis on freeware (bwa bwtsw algorithm) which yields a vaild bam file for visualization.

If it is crucial to visualize the data for you, you might ask Roche technical support to have that changed in future releases or you switch to another alignment program...
Or maybe there is a viewer somewhere else which displays insertions/deletions nicely...

[ulz_peter]

that's pretty strange that you only get one output file. Are you sure the sorting step is finished?

[litali]

I think the sorting step is finished as after all the errors I receive: net.sf.picard.sam.SortSam done. Elapsed time: 0.07 minutes.
Runtime.totalMemory()=157810688
but still, I have there only the output.bai file.
I need a file which I can upload to a viewer (maybe to uscs) so I can see the coverage of the genes in the region, including annotation, so I thought the new feature of the .BAM file would be useful as I know there are many softwares which use .BAM files...

[maubp]

I get the same error, and by visualizing the 454 Bam files it seems that instead of a substitution it declares variants as an insert and deletion, which makes absolutely no sense. But I'm afraid you can't replace the cigar string without extensive programming...

So until this has been solved you can either align your reads with different software or live with that problem...

I don't like it, but it makes perfect sense if you look at how Newbler produces ACE alignments. If you view the ACE file (e.g. in Tablet) you see the insert/deletion quite nicely (since the reference is gapped) and this caries over to the Newbler BAM file (where the reference is ungapped so the insertion isn't so clear).

Sadly the Newbler use of an insert/deletion rather than a substitution has long been a source of frustration to anyone interested in SNP detection.

Hopefully they will fix this in Newbler 2.7 given feedback - a short term solution as you say would be to edit the CIGAR strings in the SAM/BAM representation.

[maubp]

I think the sorting step is finished as after all the errors I receive: net.sf.picard.sam.SortSam done. Elapsed time: 0.07 minutes.
Runtime.totalMemory()=157810688
but still, I have there only the output.bai file.

You might find samtools easier to use than Picard (I do at least),

samtools sort example.bam example_sorted

will turn example.bam into example_sorted.bam which you then must index with:

samtools index example_sorted.bam

Perhaps that will work for you. It would be nice of Roche to have this done for you - in principle they could output a sorted BAM file.

[litali]

samtools doesn't work at all, probably something wrong with the installation. I downloaded it and extracted the file, so now i have a file :samtools-0.1.18 with many files inside. But when i tried to run samtools I recieve: bash: samtools: command not found
How this can be fixed?