Is there a way to extract the bootstrap confidence intervals for each pairwise population comparison? or atleast a record of every bootstrap replicate raw data for each so I can calculate them myself? I thought maybe the data would be logged in the .bsarchive.tsv file, but the contents of the file didn't change after running the bootstrapping, so I assumed it is the original per-locus data from the initial sampling, not the bootstrap results.
Looking at the STACKS population manual, adding --log-fst-comp to the second run (after the archive as been created) might contain the data I'm interested in, however, the job fails when I run: sbatch ./scripts3/whitelist.slurm -i ./optimize_sync_new/refmap -o ./optimize_sync_new/refmap/output_bootstrap -M ./info/synsc_popmap.tsv -W ./info/synsc_WL.tsv -X ' --fstats --fst-correction --bootstrap-fst --bootstrap-reps 1000 --log-fst-comp'
Log file with error:
populations: unrecognized option '--log-fst-comp'
populations 2.62
Usage:
populations -P dir [-O dir] [-M popmap] (filters) [--fstats] [-k [--sigma=150000] [--bootstrap [-N 100]]] (output formats)
populations -V vcf -O dir [-M popmap] (filters) [--fstats] [-k [--sigma=150000] [--bootstrap [-N 100]]] (output formats)