Hi --
Can we air feature requests here?
I noticed that star can take as input paired fastq and compressed fastq and also multiple paired fastq's - this is a very good feature of star,
but what might make star even better would be if, as input, it could accept an unmapped bam (ubam) file or multiple ubam files.
We currently handle this but...
There are many projects and organizations ( I can think of at least three big ones) that automatically make an unmapped bam file with all the read pairs, read groups and header
with metatdata as a starting point for many of their analysis pipelines and to transmit as raw data - not fastq or fastq.gz.
Is there any way to have star accept as an option an unmapped bam input, analagous to the paired fastq as an added feature? Such a feature would really really improve star from our perspective :-).
Thank you for your aligner - we really appreciate your efforts and work to build and improve this aligner for RNAseq!
Best Regards,
Daniel J McGoldrick Ph.D.
University of Washington
Department of Genome Sciences