Handle alternate contigs in GRCh38 vcf and alignment files.

[tonydisera]

Change gene.iobio to show variants and depth (from alignments) for genes on alternate contigs.

For example, gene GSTT1 for build GRCh38 (RefSeq transcripts), the app is not able to display variants (or coverage) because the gene is on an alternate contig chr22_KI270879v1_alt

Here is the message that appears:

Code to change:

• Review code in vcf.iobio.js (getChromosomes, getReferenceLengths, etc) to ensure that that alternative contigs are included along with the standard set of chromosomes in any validation.

• Review code in gene.iobio.db import script that loads the geneinfo database from Gencode and RefSeq GFF files. (Make sure that genes and transcripts on alternative transcripts are not bypassed.)

[AlistairNWard]

Should we handle alternate contigs ( eg. chr22_KI270879v1_alt)? There are thousands of them and I don't know that there are any annotations for any (many) of them anyway. If you think it's worthwhile that fine, otherwise we should just stick to 1-22, X, Y, and MT

[aaron-chapman]

MT/chrM doesn't work at all (and never has for any of my files), nor does even the most common ALS gene "C9orf72" anymore. It previously did, however (up until 'very recently' - e.g. 3-6 weeks I'd say).

[tonydisera]

@aaron-chapman, I will work on a fix for this problem ASAP.
Best,
Tony Di Sera

[tonydisera]

@aaron-chapman, the latest release of gene.iobio (version 4.11.3) has fixed the problem with mitochondrial genes. Also, genes with lower-case characters (example: C9orf72) will work. When a gene doesn't have transcripts, the app will examine gene aliases to look for a viable substitution. The app will also issue an alert if transcripts exist, but only for the alternative gene model (RefSeq vs Gencode).