#774 HBA1 descriptor calculation goes crazy, but only sometimes

[ssorgatem]

I was doing some tests with a file for a software I'm developing, and when I did certain unrelated changes, the HBA calculation went crazy.

I haven't been able to reproduce the normal behaviour, only the buggy one.

I'm using OpenBabel 2.3.1 on Debian. This happens using both pybel from OpenBabel and from cinfony 1.0.

So i've written a little script which shows what happens:

from cinfony import pybel # or just 'import pybel', the result is the same for me

testfile = 'test/test_ligands.sdf.gz'

print [mol.calcdesc(['HBA1'])['HBA1'] for mol in pybel.readfile('sdf', testfile)]
print [mol.calcdesc(['HBA2'])['HBA2'] for mol in pybel.readfile('sdf', testfile)]

When I run it it prints the following:
[35.0, 43.0, 37.0, 46.0, 37.0, 37.0]
[6.0, 6.0, 6.0, 7.0, 6.0, 6.0]

There can be seen that the HBA1 values are extremely high in comparison. The most annoying part is that I was getting sensible values from HBA1 calculation in my program, but I'm not able to reproduce them (they were values almost identical to the ones of HBA2).

Thank you,

Adrià ssorgatem@gmail.com

[ssorgatem]

These molecules get crazy HBA1 values, sometimes