check kdtree search radius as Number to include np float

[jamesmkrieger]

Otherwise, calcDistanceMatrix gives an error when cutoff=None because the maximum calculated is of type np.float32

[karolamik13]

Any particular case? When I am checking something like this:
In [4]: pp = parsePDB('8c2h.pdb').select('protein and resid 1 to 100 and name CA')
@> 9912 atoms and 1 coordinate set(s) were parsed in 0.14s.

In [5]: calcDistanceMatrix(pp, cutoff=None)
Out[5]:
array([[80.88474274, 3.86621261, 5.47263527, ..., 47.5751915 ,
50.74065399, 48.89710999],
[ 3.86621261, 80.88474274, 3.83298826, ..., 50.04154587,
53.27085114, 51.55004883],
[ 5.47263527, 3.83298826, 80.88474274, ..., 49.64807129,
52.74860382, 50.88114548],
...,
[47.5751915 , 50.04154587, 49.64807129, ..., 80.88474274,
3.82811856, 5.50502872],
[50.74065399, 53.27085114, 52.74860382, ..., 3.82811856,
80.88474274, 3.82584071],
[48.89710999, 51.55004883, 50.88114548, ..., 5.50502872,
3.82584071, 80.88474274]])

I don't have problems with obtaining the results. I installed your version, and the results are the same.

[jamesmkrieger]

It probably depends on the numpy version. I think it was in the python 3.12.6 environment. I’ll need to check

[jamesmkrieger]

It was in one of my MD simulations with measuring distances to chloride that was unbinding