Enabling the generation and use of "CSI" bam indexes

[maol-corteva]

Large genomes may have very large chromosomes (for example, the wheat genome), which are unsupported by the default .bai index files from samtools (for bam file alignments). IsoQuant 3.7 enabled its use with large genomes by allowing the passing of '-I16G' parameter to the indexing+alignment tool minimap2 (see closed issue #284)

With this capability enabled now, however, the IsoQuant.py pipeline fails at the step of indexing the minimap2 sorted BAM file. The error produced has been documented as a problem with the maximum genomic coordinate supported by the .bai index.

We need to enable the generation of .csi indexes in IsoQuant whenever the default .bai fails because of alignments beyond the max coordinate range supported by the .bai index spec.

I have added this support with a simple change to "read_mapper.py" and made a pull request for it at: #326

[andrewprzh]

Thanks a lot for the PR! I tested the patch and it seems to work just fine.