Starting with a phased (e.g. with beagle) VCF, parse into independent loci, build alignment files with reference, pull in ax/tig data, build trees, and visualize
Take a multi-locus VCF file and generate new VCFs for each locus:
perl sepLociIntoIndVCFs.pl
perl PCAandDistLocus.pl
We want to include phylogenetic trees that include:
- All of the haplotypes recovered in the CTS data
- As much A. mexicanum data as we can recover
- Data from some other A. tigrinum we have processed
We'll first generate a collection of haplotypes for each locus, and a graph that tells us which animals have which haplotypes. Unfortunately, there doesn't appear to be software that takes a phased VCF and outputs modified reference sequence haplotypes (vcf-consensus does not appear to be phasing-aware). So we'll roll our own here. In this case, we'll start with a VCF that was phased using BEAGLE (though this should work with any VCF that is vertical-bar phased). We can cull these using something like this:
perl printHaplotypes.pl /mnt/Data1/CTScombined/haplotyping/HSEM031-040.bqsr.snps.q30.passOnly.minGQ20MaxMiss50p.nuke3sparse.beaglePhased.rename.vcf haplotypeData.tsv