dupes now appearing in "G" mode

[mmaiers-nmdp]

This is a Google Colab session with these three commands

pip install py-ard

installs version 1.5.3

import pyard

ard = pyard.init()

and

ard.redux("HLA-A*02:GNF", "G")

which returns

HLA-A*02:01:01G/HLA-A*02:01:02G

We need this to be dupe squashed. Isn't this the default?

[mmaiers-nmdp]

same problem for redux mode "P" but not "lg", "lgx"

[pbashyal-nmdp]

The dupes are only "squashed" for ARD level. #333

[mmaiers-nmdp]

Let’s extend that to all resolutions. Arguably a glstring with X/X is semantically correct. I can’t think of any reason why there shouldn’t be a general dupe squashing (within allele ambiguities) within all glstring every time everywhere.