Skip to content

Latest commit

 

History

History

introgression

filetObs.py

Returns a single vector of summary stats from a window-based file as either the mean or median value. Mask option to remove statistics affected by missing data

filet_clustering.py

Splits and concatenates result from produced by FILET into direction and status

filet_clustering_mean.py

similar to above, but uses a rolling mean to merge windows instead of a strict cutoff

filet_pipeline.sh

Hard coded example of running FILET

make_filtet_mask.py

Creates masking file for use with FILET to mask training simulations

make_trainvecs_filetMP.py

python wrapper to calculate the training vector of summary statistics used with FILET using multi-processing

msMaskAllRows

https://github.com/kr-colab/FILET FILET exe to apply mask to ms-type file

normalizeTwoPopnStats.py

https://github.com/kr-colab/FILET Normalize the FILET stats vector by number of sites

removeNedOutColumnsFromMsFileKeepStars.py

https://github.com/kr-colab/FILET Remove masked columns from ms-type file, but retain * with msmove

twoPopnStats_forML

https://github.com/kr-colab/FILET FILET exe to calculate stats vector