Post date: Oct 09, 2016 6:3:11 PM
My first set of runs (with local ancestry 'Z') used L. anna (comb_lychybAutos_Anna.txt) and western L. melissa (comb_lychybAutos_MelWest.txt) populations as the source or parental populations. I want to see how robust my results are (in terms of overall patterns of ancestry freqs.) with alternative source populations. I will stick with L. anna as one parent (for now), but am using northern L. melissa (comb_lychybAutos_MelNorth.txt, 'mn' in out files) populations or L. idas (comb_lychybAutos_Idas.txt, 'id' in outfiles) populations as the other parent. I could try L. ricei as one parent later, and I need to run all the sex chromosome analyses still (with the comb_lychybSex* infiles). I am running five MCMC replicates for each data set. The results are being written to /uufs/chpc.utah.edu/common/home/u6000989/projects/lyc_hybanc/ancfreq/, but will later be moved into the mcmc sub-directory within that directory. Here is one example command (MCMC details are the same for the others):
module load gcc
module load gsl
module load hdf5
cd /uufs/chpc.utah.edu/common/home/u6000989/projects/lyc_hybanc/ancfreqs/
popanc -o /scratch/general/lustre/u6000989/outmnZBKM4.hdf5 -n 5 -d 15 -m 50000 -b 25000 -t 20 -s 1 -q 1 -z 1 infiles/comb_lychybAutos_Anna.txt infiles/comb_lychybAutos_MelNorth.txt infiles/pntest_lychybAutos_BKM.txt
cp /scratch/general/lustre/u6000989/outmnZBKM4.hdf5 ./