This script combines (and optionally filters) results from the Keeneland branch of Beast-Beagle.
Put the script in the same directory as your output files, and run it with the following arguments:
$ post.pl prefix numRanks stateIncrement discardField distance
prefix - Look for Beagle output files named prefix + '' + rank#.log - e.g. prefix0.log, prefix_1.log, ...
nRanks - the number of mpi ranks used in the run
stateIncrement - the frequency with which states were saved to file, used in relabeling states
nStates - the number of entries to consisder per file, 0 = consider all
discardField - (Optional) discard based on this field in the output file e.g. 1 - posterior, 2 - prior, 3 - likelihood