From: Raphael A. Bauer <raphael.bauer@go...> - 2008-10-01 12:50:11
my propositoins to the PDBReader and PDBWriter are now as branch
in the svn. I also added a testcase and a file that represents a
binding site with mixed amino acids
and hetero atoms (binding_site.pdb). All other testcases are solved
Some main points and ideas:
- now writes out all PDBAtom information (residuename instead of MOL,
- is NOT able to write out models or chains or so. it is a stupid
method, but does it's job for me at least.
- various general bugfixes (lengths, connection informations etc...)
Something for discussion:
- if you read in a simple "mixed" pdb file containing atoms and
hetatoms the PDBReader is not able to read it
because it assumes the keyword "PROTEIN" and so only reads in ATOM
records and no HETATOM records.
=> I added a method called setLazyMode(boolean) that manually switches
on the protein mode and
solves this thing.
- I also added a method setParseHetAtoms(boolean) that omits parsing
of hetero atoms for "performance" reasons.
What do you think about those two methods, and or solutions to the
"problem". In fact I can not judge if
it makes sense to change the class signature for these reasons.
From: Egon Willighagen <egon.willighagen@gm...> - 2008-10-08 12:48:28
I still need to check your patches in detail, but...
On Wed, Oct 1, 2008 at 2:50 PM, Raphael A. Bauer
> => I added a method called setLazyMode(boolean) that manually switches
> on the protein mode and
> solves this thing.
> - I also added a method setParseHetAtoms(boolean) that omits parsing
> of hetero atoms for "performance" reasons.
These two should be handled as IOSettings... see this CDK News 1.2 for
some information on it.