Hello,
Thank you for your explanation.
> On Jan 25, 2016, at 8:05 PM, Geoffrey Hutchison <geoff.hutchi...@gmail.com>
> wrote:
>
>> As it is shown here, the ring is recognized as an aromatic ring and so the
>> carbons as “ca”.
>> However, based on the GAFF force field, C1 and C4 must be “cc” , C2 and C3
>> must be “cd”.
>
> Weird that GAFF doesn't think furan is aromatic.
>
>> Another problem is that oxygen (atom number 9) is not recognized at all.
>
> Well, that depends on the SMARTS you're using for oxygen typing. Without
> seeing your SMARTS patterns, I'd guess you don't have any fallback patterns.
Actually, the atom types I sent you through the previous email was generated
based on the original gaff.prm implemented in openbabel.
I am defining new SMARTS patterns to get correct GAFF atom types for different
molecules, for instance this is the correct output for furan.
A T O M T Y P E S
IDX TYPE RING
1 cc AR
2 ha NO
3 cd AR
4 ha NO
5 cd AR
6 ha NO
7 cc AR
8 ha NO
9 os AR
As you mentioned, openbabel goes over gaff.prm from simple patterns to the
complex ones. But, it seems the problem is to distinguish aliphatic conjugated
rings from aromatic rings, particularly for 5- and 6- membered rings. Because,
it uses “ca” atom types very often for aliphatic conjugated ring systems rather
than “cc” or “cd”.
Therefore, even if we have correct SMARTS pattern for aliphatic conjugated
rings, it does not use them because of the failure of the matching process.
Cheers,
Mohammad
>
>> My question is that which function or routine in the code recognizes
>> aromatic rings from non-aromatic rings?
>
> There are multiple routines in aromatic perception and Kekulization. But you
> don't really need to get involved in that (nor am I going to accept any
> patches for GAFF typing that change aromaticity detection).
>
> What you want is a set of SMARTS patterns going from more general to more
> specific. Take a look at data/UFF.prm for a very simple example. So
> *everything* matches the first pattern, but it might get supplanted by a more
> specific example:
>
> atom [#6] C_3 Generic sp3 C
> atom [C^2] C_2 sp2 non-aromatic C
> atom [C^1] C_1 sp hybridized C
> atom [c] C_R aromatic C
> atom [#6H1D3r5] C_R aromatic C (cyclopentadienyl ring)
>
> Notice that for most atom typing in UFF.prm, we use #6 for carbon -
> regardless of whether it's an aromatic or aliphatic carbon.
>
> My guess is that the oxygen in furan doesn't have a SMARTS pattern that
> matches [#8] or [o] (the latter being an aromatic oxygen).
>
> If you want to "correct" carbons in a furan to "cc" or "cd", then construct
> SMARTS patterns for them:
> atom [oD2] o "sp2 O - correct for aromatic oxygen"
>
> I *think* that moving the "atom [c] ca "pure aromatic atom"
> before the various conjugated patterns would work - you'd be able to replace
> a generic aromatic atom with more specific atoms. You could also have
> something like:
>
> atom [cR5]o cc "correct conjugated carbon in furans"
>
> Hope that helps,
> -Geoff
------------------------------------------------------------------------------
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
_______________________________________________
OpenBabel-Devel mailing list
OpenBabel-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-devel
