General pipeline follows from Browning et al. 2018
- align reads with bwa-mem2 (to do)
- call SNPs w/ freebayes
- filter with vcftools
- phase/impute with Beagle5
- identify IBD with refinedIBD
- merge IBD segments with mergeIBD
- estimate Ne over time with IBDNe
- plot results (to do)
Requires estimation of small IBD regions. This means careful genotype filtering is critical, but also requires high enough SNP density to pick up small regions. And means you need a high resolution genetic map.
- process substitution mumbo jumbo in freebayes parallel fail on zsh and maybe other shells. use bash.
- Currently memory and CPU are hard-coded into shell commands within each rule, and are also specified in the file
submit.yaml. Make sure these match. - Need to run more bootstrap replicates for any final analysis of IBDNe.
#Relate subdirectory
Also trying with relate.