From: JC Gumbart (gumbart_at_physics.gatech.edu)
Date: Mon May 30 2016 - 07:19:44 CDT
Fair enough - although I would suggest that a direct comparison using a relatively inaccurate force field may not be as useful as you think.
Regarding your question, it’s too abstract for me to understand. Computers are stupid - they will do exactly what we tell them and no more. If there are four atoms with distinct atom types, then the simulation program will expect four parameters. It won’t guess anything unless you’ve programmed it to do so.
Best,
JC
> On May 29, 2016, at 12:12 PM, Benjamin Hill <ben.luc.chem_at_gmail.com> wrote:
>
> Thank you for your reply.
>
> There is a reason why I wish to use Charmm27 -- This regards a preliminary study regarding different NTPs, performed by a group of others who ran some simulations using charmm27 -- so I want to be consistent at this stage. Simulations will be re-run in charmm36 when needed.
>
> The following is more of a technical question.
>
> Suppose you have 1,3-cyclobutene. Say all 4 carbons are flat, on the same plane.. Say the carbon atom names are A,B,C,D.
> Say the the 4 bond angles in this square structure are a,b,c,d.
>
> The computer wants to know the bond angles between these carbons. As long as you tell it bond angles a,b,c .... then it can just calculate d.
>
> If I now modify 1,3-cyclobutene, by (for example) replacing one -H with an -OH group, then might this change how the computer sees the molecule?
> -- That is to say, might the computer now need to know bond angles b,c,d angles (--then itself calculating 'a'?), instead of the angles like it was before -- needing to know angles a,b,c and then calculating 'd'?
>
> -- In this analogy, is it possible that only all three bond angles a,b,c are defined in the parameter set, but not all three of b,c,d bond angles?
>
>
> -- Ultimate question: At what stage is an error occurring in this process?
>
>
> Thank you,
> BH
>
>
>
>
>
> On Fri, May 27, 2016 at 3:28 PM, Lennart Nilsson <Lennart.Nilsson_at_ki.se <mailto:Lennart.Nilsson_at_ki.se>> wrote:
> I agree with JC, there would have to be a very strong reason to use an old version of a force field; version 36 for nucleic acids has been available for about four years now.
>
> I suspect that the cause of the problem is a mixup of atom types for ribose and deoxyribose. ATP from the distributed files works, but the new ITP does not, so a mistake was made in constructing ITP. The devil is, as usual, in the details. <>
>
>
> Best,
>
> Lennart
>
>
>
> From: owner-namd-l_at_ks.uiuc.edu <mailto:owner-namd-l_at_ks.uiuc.edu> [mailto:owner-namd-l_at_ks.uiuc.edu <mailto:owner-namd-l_at_ks.uiuc.edu>] On Behalf Of JC Gumbart
> Sent: den 27 maj 2016 21:14
> To: Benjamin Hill <ben.luc.chem_at_gmail.com <mailto:ben.luc.chem_at_gmail.com>>
> Cc: namd-l <namd-l_at_ks.uiuc.edu <mailto:namd-l_at_ks.uiuc.edu>>
> Subject: Re: namd-l: No parameter for ribose ring, Charmm 27
>
>
>
> I just checked an old version of the par_all27_na file and also didn’t find it. It might be in one of the old stream files then? I imagine someone at some point addressed this - maybe search the CHARMM forums if you’re determined to use it.
>
>
>
> To be honest though, I’m at the point personally where I’m going to start rejecting papers outright that refuse to use updated force fields, which have been proven to be significantly better in a variety of respects. Why not take advantage of the great work Lennart has done to make your simulations better?
>
>
>
> Best,
>
> JC
>
>
>
> On May 27, 2016, at 6:51 AM, Benjamin Hill <ben.luc.chem_at_gmail.com <mailto:ben.luc.chem_at_gmail.com>> wrote:
>
>
>
> Yes... but then why does ATP work and not my molecule? Here, both have same atom types.
>
> I would not have to upgrade to charmm36 to run a simulation with ATP... so why with ITP? No difference in ribose ring atom names or types.
>
> Please help!!
> Thank you!!
>
> On May 26, 2016 6:42 PM, "JC Gumbart" <gumbart_at_physics.gatech.edu <mailto:gumbart_at_physics.gatech.edu>> wrote:
>
> C2’ is type CN7B in CHARMM36. You should update your model using the latest force field anyway, which will probably also solve your issue.
>
> Best,
> JC
>
> > On May 26, 2016, at 6:18 PM, Benjamin Hill <ben.luc.chem_at_gmail.com <mailto:ben.luc.chem_at_gmail.com>> wrote:
> >
> > Hello,
> > I am trying to make topology file for inosine triphosphate (ITP). ITP has a very similar structure to ATP. Long story short... I used ATP as a structural template. I remove the adenine "group" and attach the new "group" to create ITP... and prepared the new ITP topology file (again, using ATP as a template -- I removed the adenine group and attached the new one to make ITP).
> > When I run the simulation, I get a weird error -- it says it' can't identify certain parameters for parts of the ribose ring of ITP -- even though I did not touch this section of the ATP template in order to create the ITP. For instance, it says it' can't angle parameters for atoms C4' - C3' - C2' which are atom types CN7-CN7-CN7. Indeed it is true -- there are no such parameters in the parameter file for this angle. But this is for ribose. When I run ATP simulation it is fine, but not for ITP.
> > What is going on?
> >
> > Here is my ITP topology file (residue name is ITX)
> > -- Below it is a copy of a PDB for an ITP molecule
> >
> >
> > RESI ITX -4.00 !
> > !
> > !
> > GROUP !
> > ATOM C4' CN7 0.16 !
> > ATOM H4' HN7 0.09 !
> > ATOM O4' ON6B -0.50 ! O6 (N6, H61, H62 removed)
> > ATOM C1' CN7B 0.16 ! ||
> > ATOM H1' HN7 0.09 ! NEW for ITP-> H1 C6
> > GROUP ! \ / \
> > ATOM C5 CN5G 0.00 ! N1 C5--N7\\
> > ATOM N7 NN4 -0.60 ! | || C8-H8
> > ATOM C8 CN4 0.25 ! C2 C4--N9/
> > ATOM H8 HN3 0.16 ! / \\ / \
> > ! H2 N3 \
> > ! \
> > ATOM N1 NN2G -0.32 ! \
> > ATOM H1 HN2 0.26 !!!! NEW FOR ITP |
> > ATOM N9 NN2B -0.01 ! |
> > ! |
> > ! |
> > ATOM O6 ON1 -0.51 ! \
> > ATOM C2 CN4 0.50 ! \
> > ATOM H2 HN3 0.08 ! (-)O3G O2B O1A H5' H4' O4' \
> > ATOM N3 NN3I -0.61 ! | | | | \ / \ |
> > ATOM C4 CN5 0.26 !O1G=PG-O3B-PB-O3A-PA-O5'-C5'---C4' C1'
> > ATOM C6 CN1 0.54 ! | | | | \ / \
> > ! (-)O2G (-)O1B (-)O2A H5'' C3'--C2' H1'
> > ! / \ / \
> > ! O3' H3' O2' H2''
> > ! | |
> > ! H3T H2'
> > !
> > !
> > GROUP !
> > ATOM C2' CN7 0.14 !
> > ATOM H2'' HN7 0.09
> > ATOM O2' ON5 -0.66
> > ATOM H2' HN5 0.43
> > GROUP
> > ATOM C3' CN7 0.14
> > ATOM H3' HN7 0.09
> > ATOM O3' ON5 -0.66
> > ATOM H3T HN5 0.43
> > GROUP
> > ATOM C5' CN8 -0.08
> > ATOM H5' HN8 0.09
> > ATOM H5'' HN8 0.09
> > ATOM O5' ON2 -0.62
> > ATOM PA P 1.50
> > ATOM O1A ON3 -0.82
> > ATOM O2A ON3 -0.82
> > ATOM O3A ON2 -0.74
> > ATOM PB P2 1.50
> > ATOM O1B ON3 -0.82
> > ATOM O2B ON3 -0.82
> > ATOM O3B ON2 -0.86
> > ATOM PG P2 1.10
> > ATOM O1G ON3 -0.90
> > ATOM O2G ON3 -0.90
> > ATOM O3G ON3 -0.90
> >
> > BOND C6 N1
> > BOND O5' C5' O5' PA PA O1A PA O2A PA O3A
> > BOND O3A PB PB O1B PB O2B PB O3B O3B PG
> > BOND PG O1G PG O2G PG O3G
> > BOND C5' C4' C4' O4' C4' C3' O4' C1'
> > BOND C1' N9 C1' C2' N9 C4 N9 C8 C4 N3
> > BOND C2 N1
> > BOND N1 H1
> > BOND C6 C5 C5 N7
> > BOND C2' C3' C2' O2' O2' H2' C3' O3' O3' H3T
> > BOND C1' H1' C2' H2'' C3' H3' C4' H4' C5' H5'
> > BOND C5' H5'' C8 H8 C2 H2
> > DOUBLE C6 O6
> > DOUBLE N3 C2
> > DOUBLE C4 C5
> > DOUBLE N7 C8
> > IMPR C6 N1 C5 O6
> > IMPR C2 N1 N3 H2
> >
> > PATC FIRS NONE LAST NONE
> >
> >
> >
> >
> > ############
> > HERE IS THE ITP PDB:
> > ############
> >
> > ATOM 1 PA ITX A 426 57.691 -8.399 58.192 1.00 0.00 ITX P
> > ATOM 2 PB ITX A 426 59.580 -6.656 58.876 1.00 0.00 ITX P
> > ATOM 3 PG ITX A 426 60.579 -6.580 56.252 1.00 0.00 ITX P
> > ATOM 4 C5' ITX A 426 56.437 -10.527 59.062 1.00 0.00 ITX C
> > ATOM 5 O5' ITX A 426 57.089 -9.297 59.376 1.00 0.00 ITX O
> > ATOM 6 C4' ITX A 426 55.091 -10.635 59.927 1.00 0.00 ITX C
> > ATOM 7 O4' ITX A 426 55.321 -10.831 61.277 1.00 0.00 ITX O
> > ATOM 8 C3' ITX A 426 54.086 -9.466 59.815 1.00 0.00 ITX C
> > ATOM 9 O3' ITX A 426 52.715 -9.964 59.922 1.00 0.00 ITX O
> > ATOM 10 C2' ITX A 426 54.537 -8.666 61.056 1.00 0.00 ITX C
> > ATOM 11 O2' ITX A 426 53.508 -7.808 61.601 1.00 0.00 ITX O
> > ATOM 12 C1' ITX A 426 54.925 -9.710 62.048 1.00 0.00 ITX C
> > ATOM 13 N1 ITX A 426 55.869 -8.550 66.999 1.00 0.00 ITX N1
> > ATOM 14 O1A ITX A 426 58.788 -9.190 57.584 1.00 0.00 ITX O
> > ATOM 15 O1B ITX A 426 59.543 -5.585 59.959 1.00 0.00 ITX O
> > ATOM 16 O1G ITX A 426 61.649 -5.564 56.369 1.00 0.00 ITX O
> > ATOM 17 O6 ITX A 426 58.050 -8.264 66.990 1.00 0.00 ITX O
> > ATOM 18 C2 ITX A 426 54.727 -8.711 66.355 1.00 0.00 ITX C
> > ATOM 19 O2A ITX A 426 56.645 -7.978 57.217 1.00 0.00 ITX O
> > ATOM 20 O2B ITX A 426 60.459 -7.891 59.050 1.00 0.00 ITX O
> > ATOM 21 O2G ITX A 426 60.884 -8.050 56.516 1.00 0.00 ITX O
> > ATOM 22 N3 ITX A 426 54.536 -9.112 65.115 1.00 0.00 ITX N
> > ATOM 23 O3A ITX A 426 58.048 -7.150 58.981 1.00 0.00 ITX O
> > ATOM 24 O3B ITX A 426 59.553 -6.040 57.378 1.00 0.00 ITX O
> > ATOM 25 O3G ITX A 426 59.755 -6.560 54.929 1.00 0.00 ITX O
> > ATOM 26 C4 ITX A 426 55.744 -9.158 64.424 1.00 0.00 ITX C
> > ATOM 27 C5 ITX A 426 56.989 -9.002 64.967 1.00 0.00 ITX C
> > ATOM 28 C6 ITX A 426 57.063 -8.657 66.351 1.00 0.00 ITX C
> > ATOM 29 N7 ITX A 426 57.989 -9.178 63.962 1.00 0.00 ITX N
> > ATOM 30 C8 ITX A 426 57.265 -9.373 62.852 1.00 0.00 ITX C
> > ATOM 31 N9 ITX A 426 55.929 -9.362 63.062 1.00 0.00 ITX N
> > ATOM 32 H1 ITX A 426 56.001 -8.211 68.403 1.00 0.00 ITX H
> > ATOM 34 H2 ITX A 426 53.836 -8.527 66.954 1.00 0.00 ITX H
> > ATOM 35 H8 ITX A 426 57.736 -9.480 61.874 1.00 0.00 ITX H
> > ATOM 36 H1' ITX A 426 54.065 -10.145 62.601 1.00 0.00 ITX H
> > ATOM 37 H2' ITX A 426 52.720 -8.343 61.724 1.00 0.00 ITX H
> > ATOM 38 H3' ITX A 426 54.104 -8.758 58.959 1.00 0.00 ITX H
> > ATOM 39 H4' ITX A 426 54.622 -11.596 59.627 1.00 0.00 ITX H
> > ATOM 40 H5' ITX A 426 57.007 -11.421 59.392 1.00 0.00 ITX H
> > ATOM 41 H3T ITX A 426 52.809 -10.870 59.617 1.00 0.00 ITX H
> > ATOM 42 H2'' ITX A 426 55.477 -8.130 60.805 1.00 0.00 ITX H
> > ATOM 43 H5'' ITX A 426 56.271 -10.685 57.975 1.00 0.00 ITX H
> > END
>
>
>
>
This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:22:13 CST