Hello,
I’m using openbabel to convert a bunch of old Chemdraw files into SMILES
strings for further processing, and I came across a pair of what appear to be
regressions in the chemdraw structure handling that I was curious if anyone
else had run into.
Many of the files have fairly standard chemical abbreviations in them, such
as SO3 for a sulfate or N3 for an azide or COOCH2CH3 for an ethyl ester. When
I converted the files to SMILES using 2.3.2, these abbreviations were mostly
brought along and correctly parsed (the charges on the azide were absurd, but
easily set right with sed). But now, when I try the same conversions in the
latest release from github (2.4.90, it appears), on these molecules, I get an
asterisk and an error message that the alias was not chemically interpreted.
It seems like this is a loss of functionality from the earlier version. Is
there a way to get the earlier behavior back, or have I stumbled over a dark
corner where I was not meant to go?
Also, the chemdraw files are not named with the name of the molecule, but
rather with an obscure numbering system. Nevertheless, the name of the
molecule being extracted is present in the file, and it gets output with the
SMILES string in 2.3.2. Once again, this functionality seems to be lost in
2.4.90. Is there an option to restore the behavior, or should I just grin and
keep using 2.3.2?
Finally, as a last question, is there a good way to tell openbabel to only
process the largest molecule present? I have a lot of counterions that are
coming through that I would gladly ignore for this project.
Thanks,
Andy
--
Andrew Fant | Systems Administrator
[email protected] | Lei Shi Lab , NIH/NIDA/IRP
(443)740-2849 |
_______________________________________________
OpenBabel-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
