Thank you for your comment. It appears that the problem is coming more from
the charges that from the bond. The problematic type was coming from an O
in a carboxylate moiety.
* starting from 10gs_ligand.xyz (copied below), I converted it to sdf with
"obabel 10gs_ligand.xyz -O 10gs_ligand_NEW1.sdf". The types found are the
same as in the .xyz file (=problematic). The bond perception is the same as
in the original sdf file, however the atoms 4 and 9 (the two oxygens of the
carboxylate) have an assigned charge of 0.
* if I do "obabel 10gs_ligand.xyz -O 10gs_ligand_NEW2.sdf -p 7", then both
the charges and the types are assigned correctly as in the original sdf.
(note that both new .sdf files have the same number of atoms).
Nicolas
*********************
59
10gs_ligand
N 15.08800 10.79800 23.54700
C 15.01000 9.98700 24.79200
C 16.11500 8.92400 24.83000
O 16.52000 8.51500 25.94000
C 13.63500 9.32700 24.90800
C 13.39400 8.70800 26.27100
C 12.04500 8.04600 26.40200
O 11.29300 7.93600 25.43500
O 16.57800 8.52400 23.74400
N 11.72600 7.64200 27.62800
C 10.47200 6.96700 27.93400
C 10.72600 5.48400 28.20600
S 11.29100 4.52400 26.81000
C 9.72900 3.80400 26.26200
C 8.93000 3.17100 27.37000
C 7.64000 3.61400 27.65000
C 9.46400 2.13500 28.13300
C 6.89300 3.03700 28.67300
C 8.72300 1.55000 29.16100
C 7.43700 2.00100 29.43000
C 9.83400 7.55000 29.18000
O 10.52200 8.02300 30.08400
N 8.51200 7.46800 29.22900
C 7.74000 7.93300 30.36600
C 6.55500 7.06200 30.63300
C 5.33000 7.31500 30.02700
C 6.68300 5.94100 31.44100
C 4.25000 6.45900 30.22000
C 5.61100 5.08100 31.64000
C 4.39200 5.33900 31.02700
C 7.45200 9.43300 30.35400
O 7.11600 9.95700 31.43300
O 7.56900 10.06800 29.28400
H 14.35220 11.48700 23.54820
H 14.98240 10.19510 22.74610
H 15.98210 11.26140 23.50330
H 15.14780 10.65930 25.65170
H 13.55820 8.53820 24.14520
H 12.86280 10.08910 24.72650
H 13.46830 9.49980 27.03090
H 14.17190 7.95160 26.45220
H 12.37410 7.80850 28.37110
H 9.78640 7.07760 27.08100
H 9.78470 5.04000 28.56220
H 11.48780 5.41130 28.99610
H 9.94780 3.03330 25.50830
H 9.12100 4.59980 25.80700
H 7.21180 4.41890 27.06360
H 10.46670 1.77900 27.92590
H 5.89030 3.39270 28.88040
H 9.15020 0.74530 29.74830
H 6.85800 1.54820 30.22680
H 8.02700 7.06940 28.45060
H 8.39770 7.78750 31.23560
H 5.21460 8.18930 29.39670
H 7.63170 5.73410 31.92290
H 3.29990 6.66660 29.74110
H 5.72570 4.20880 32.27350
H 3.55430 4.66790 31.17830
2015-07-09 19:37 GMT-04:00 Geoffrey Hutchison <geoff.hutchi...@gmail.com>:
> * with ./Typing.exe 10gs_ligand.sdf, the 4th atom is of type "o" (which is
> the good type)
> * with ./Typing.exe 10gs_ligand.xyz, the 4th atom is of type "os"
>
>
> While I can't comment on the particular files, an XYZ file has no bonding
> information whatsoever. While the bond perception in OB is fairly good,
> there's no way that typing can be perfect when you've thrown away all the
> bonding information.
>
> So yes. Atom types for GAFF or other force fields will depend on the
> bonding information, which depend on the file type.
>
> Where possible, use file types like SDF or Mol2 that retain bonding
> information if you want to retain atom typing.
>
> -Geoff
>
>
------------------------------------------------------------------------------
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide you with the tools and support that
you need to offload your IT needs and focus on growing your business.
Configured For All Businesses. Start Your Cloud Today.
https://www.gigenetcloud.com/
_______________________________________________
OpenBabel-discuss mailing list
OpenBabel-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
