I am trying to run a genome-guided Trinity assembly with various RNA-seq lanes that I have mapped to my genome, merged the BAM files, and sorted with samtools. I have uploaded this BAM file into the Galaxy interface for running genome-guided Trinity, but it is requiring me to also upload FASTQ files, even though this shouldn’t be required for genome-guided Trinity, for instance when running on the command line. Is there a way to bypass the requirement for the extra FASTQ files, since the BAM file should have all the relevant info for genome-guided assembly?
In the Galaxy wrapped version of Trinity, distinct input fastq reads are required – even when using the option “Use the genome guided mode?” as “Yes” with mapped BAM inputs.
I’ve asked the tool wrapper authors what they think about using the BAM data directly. It may be an enhancement already under consideration. They may reply here or at Gitter – and please feel free to join in the discussion at either: https://gitter.im/galaxy-iuc/iuc?at=5ec453ef95576839186d4311
Meanwhile, you’ll need to supply the fastq reads plus the bam data as an input.
Thanks for your response! I’ll run it with the FASTQ and BAM for now, but would be nice to be able to supply the BAM only - I’m not sure but it may also improve run time/memory usage.