OK, this seems to work.... so here's a description of my patch.
I edited "include/openbabel.forcefield.h", adding a simple function:
"OBBitVec &GetFixedBitVec() { return _fixed; }"
I edited "src/forcefield.cpp", adding this line:
rl.SetFixAtoms(_constraints.GetFixedBitVec());
to each of the following three functions, immediately preceding the
call to "rl.Setup(_mol);":
OBForceField::SystematicRotorSearchInitialize
OBForceField::RandomRotorSearchInitialize
OBForceField::WeightedRotorSearch
And the conformer search routines now appear to respect fixed atom
constraints. :)
Hopefully this is an appropriate place to post such modifications.
Let me know what you think.
Cheers,
Chris
On Tue, Jun 29, 2010 at 12:14 PM, Christopher Wassman
<cwass...@ics.uci.edu> wrote:
> Here's a bit more info, perhaps useful...
>
> For whatever reason
>
> Inside of
> OBRotorList::FindRotors
>
> The call to HasFixedAtoms() returns false, even though I've previously
> specified constraints to the setup of the force field.
>
> So... it seem that _fix is empy (HasFixedAtoms() is declared in
> include/openbabel/rotor.h)
>
> Seems _fix is not in sync with _constraints._fixed
>
> Anyway, it would seem that the rotor search codes should be calling
> OBRotorList.SetFixAtoms(...) specifying the _fixed OBBitVec from
> _constraints... but I can't find any where they are doing so.
>
> At least that's my naive take on it. I must admit my C++ is very
> rusty and I've just started looking at this code base last week, so
> take this with the appropriate caution. :)
>
> Cheers,
> Chris.
>
> On Mon, Jun 28, 2010 at 7:43 AM, Noel O'Boyle <baoille...@gmail.com> wrote:
>> I've never used OBConstraints, but just as general advice: when
>> setting up the forcefield, you should ensure that (a) the atoms have
>> 3D coordinates, (b) hydrogens have already been added, and (c) that
>> calling the Setup function is successful.
>>
>> // We pass the constraints as argument for Setup()
>> if (!pFF->Setup(mol, constraints)) {
>> cerr << "ERROR: could not setup force field." << endl;
>> }
>>
>> - Noel
>>
>> On 25 June 2010 01:03, Christopher Wassman <cwass...@ics.uci.edu> wrote:
>>> Tried to implement fixing atoms as part of a conformational search....
>>> as a straightforward
>>> modification of the 'obconformer' tool.... but perhaps I'm missing
>>> something. I'm using the source from
>>> openbabel-2.2.3
>>>
>>>
>>> I simply added this (in the "while(ifs.peek() != EOF && ifs.good())" loop):
>>>
>>> OBFFConstraints constraints;
>>> int i = 0;
>>> for (i=1; i<=3; i++) {
>>> cout << " Fixing atom index " << i << "...\n";
>>> OBAtom *atom;
>>> atom = mol.GetAtom(i);
>>> if (!atom->IsHydrogen())
>>> {
>>> constraints.AddAtomConstraint(i);
>>> }
>>> else
>>> {
>>> cerr << "Error! heavy atoms first !" << endl;
>>> }
>>> }
>>>
>>> pFF->Setup(mol,constraints);
>>> pFF->WeightedRotorSearch(weightSteps, geomSteps);
>>> ...
>>> etc as in "tools/obconfomer.cpp"
>>>
>>>
>>> and removed the old "pFF->Setup(mol);" statement
>>>
>>> I tried this on a simple butane molecule (defined in a pdb), where the
>>> first atoms listed are the carbon backbone, in linear order.
>>>
>>> I'd expect this to "freeze" the first 3 carbon atoms, however, in the
>>> output only 2 of the 4 carbon atoms are held in fixed position. (The
>>> first and the last carbons moved.)
>>>
>>> Trying this approach to freeze all 4 of the carbon atoms, results in a
>>> conformer where only 1 of the carbon atoms have moved. (The first
>>> carbon moved.)
>>>
>>> Anyone able to shed some light on this for me? What am I missing?
>>>
>>> ------------------------------------------------------------------------------
>>> ThinkGeek and WIRED's GeekDad team up for the Ultimate
>>> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
>>> lucky parental unit. See the prize list and enter to win:
>>> http://p.sf.net/sfu/thinkgeek-promo
>>> _______________________________________________
>>> OpenBabel-discuss mailing list
>>> OpenBabel-discuss@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>>>
>>
>>
>
------------------------------------------------------------------------------
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
_______________________________________________
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
