Eric Banks 3e7714629f Scrapped the whole idea of an int/long as an index into the ACset: with lots of alternate alleles we run into overflow issues. Instead, simply use the ACcounts array as the hash key since it is unique for each AC conformation. To do this, it needed to be wrapped inside an object so hashcode() would work.		2011-12-08 23:50:54 -05:00
..
config	Reorganized the codebase beneath top-level public and private directories,	2011-06-28 06:55:19 -04:00
src	Scrapped the whole idea of an int/long as an index into the ACset: with lots of alternate alleles we run into overflow issues. Instead, simply use the ACcounts array as the hash key since it is unique for each AC conformation. To do this, it needed to be wrapped inside an object so hashcode() would work.	2011-12-08 23:50:54 -05:00
test	Need to use longs for the set index (because we can run out of ints when there are too many alternate alleles). Integration tests now use the multiallelic implementation.	2011-12-08 15:31:02 -05:00