VMD-L Mailing List
From: JC Gumbart (gumbart_at_physics.gatech.edu)
Date: Tue Oct 24 2017 - 17:21:14 CDT
- Next message: Seibold, Steve Allan: "Wrapping molecules in unit box"
- Previous message: Chitrak Gupta: "Re: write a new dcd file for a sub selection of atoms"
- Maybe in reply to: JC Gumbart: "Re: FFTK Dihedral Optimization Error"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
I had a look at your structure. The dihedral angles that fail all involve three ring atoms. Did the scan run a few steps before failing? My suggestion would be to shrink the scan size (not sure what an appropriate range will be, but you can see where the modredundant scan crashed) and do it in smaller increments (maybe 5 deg. instead of 15 deg.). Also, I bet if you visualize the output of your cartesian scan, it probably looks pretty weird.
Best,
JC
> On Oct 21, 2017, at 12:42 PM, Nikolaos Galanakis <ng1807_at_nyu.edu> wrote:
>
> Many thanks for your reply.
>
> I have in total 15 dihedral angles in my structure. For 12 of them, gaussian performs the scan without any problems.
> But for three of them the scan fails. My structure is this one:
>
> C1 1.0360000133514404 0.8849999904632568 0.03500000014901161
> O2 2.1600000858306885 1.6710000038146973 0.05999999865889549
> C3 -0.23800000548362732 1.3890000581741333 -0.23600000143051147
> H4 -0.37299999594688416 2.447000026702881 -0.460999995470047
> C5 -1.3380000591278076 0.5299999713897705 -0.2370000034570694
> H6 -2.318000078201294 0.9240000247955322 -0.4970000088214874
> C7 -1.1699999570846558 -0.8339999914169312 0.029999999329447746
> C8 -2.302000045776367 -1.809999942779541 0.027000000700354576
> C9 0.11800000071525574 -1.3309999704360962 0.27799999713897705
> H10 0.23899999260902405 -2.3970000743865967 0.453000009059906
> C11 1.215000033378601 -0.47999998927116394 0.2879999876022339
> H12 2.2170000076293945 -0.8479999899864197 0.4860000014305115
> H13 1.902999997138977 2.5880000591278076 -0.14499999582767487
> O14 -2.1730000972747803 -2.9830000400543213 -0.3050000071525574
> N15 -3.500999927520752 -1.246000051498413 0.4429999887943268
> N16 -4.622000217437744 -2.0339999198913574 0.5320000052452087
> H17 -3.453000068664551 -0.36899998784065247 0.9559999704360962
> C18 -5.675000190734863 -1.4839999675750732 1.0509999990463257
> H19 -7.2270002365112305 -2.444999933242798 2.183000087738037
> C20 -5.738999843597412 -0.06499999761581421 1.555999994277954
> C21 -6.900000095367432 -2.3420000076293945 1.1419999599456787
> H22 -7.730000019073486 -1.8949999809265137 0.5820000171661377
> H23 -6.684000015258789 -3.3310000896453857 0.7350000143051147
> H24 -5.072999954223633 0.08699999749660492 2.4159998893737793
> H25 -5.442999839782715 0.6449999809265137 0.7739999890327454
> H26 -6.751999855041504 0.1860000044107437 1.875
>
> and the angles that fail to scan in gaussian are: 2-1-3-5, 2-1-11-9 and 3-5-7-8.
>
>
>
> On Sat, Oct 21, 2017 at 12:40 PM, Nikolaos Galanakis <ng1807_at_nyu.edu <mailto:ng1807_at_nyu.edu>> wrote:
> Many thanks for your reply.
>
> I have in total 15 dihedral angles in my structure. For 12 of them, gaussian performs the scan without any problems.
> But for three of them the scan fails. My structure is this one:
>
>
>
> On Fri, Oct 20, 2017 at 5:54 PM, JC Gumbart <gumbart_at_physics.gatech.edu <mailto:gumbart_at_physics.gatech.edu>> wrote:
> Can you elaborate a bit? Is it all the angles or just some of them, and if so, which ones?
>
> I don’t think there’s any way to do it with cartesian coordinates based on our method of fitting (or at least I never seriously considered it).
>
> Best,
> JC
>
> > On Oct 20, 2017, at 12:50 PM, Nikolaos Galanakis <ng1807_at_nyu.edu <mailto:ng1807_at_nyu.edu>> wrote:
> >
> > Hi!
> >
> > I am trying to optimize dihedral force field parameters for a structure using FFTK.
> > When I run the gaussian to scan for the dihedral angles with the modredundant option I get an error (due to the modredundant geometry) so I scan using cartesian geometry which is performed correctly. However, when I try to run the optimization in FFTK using these gaussian log files I get the error "can't read scanDihDef: no such variable".
> > Is there any way to make VMD read correctly the gaussian log files resulting from dihedral scan with cartesian geometry otpion?
> >
> > Many thanks for your time
> >
> > Nikos
>
>
>
- Next message: Seibold, Steve Allan: "Wrapping molecules in unit box"
- Previous message: Chitrak Gupta: "Re: write a new dcd file for a sub selection of atoms"
- Maybe in reply to: JC Gumbart: "Re: FFTK Dihedral Optimization Error"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]