[gmx-users] Re: improper dihedral angle in charmm and opls aa (Mark Abraham)

Mark Abraham Mark.Abraham at anu.edu.au
Sun Feb 12 01:58:19 CET 2012


On 12/02/2012 9:22 AM, Tom wrote:
> Dear Mark and Gromacs Users,
> Can you be more specific in explanation about improper dihedral of 
> charmm in gromacs?
> For opls aa, it is very clear that every item of improper dihedral 
> angle is described once
> on the *rtp file.
> I am quite confused about the one in the charmm.
> Ihe same angle is sometimes described twice despite of the difference 
> in the atoms' sequence.

Like I said last time, different sequence of the same four atoms means 
an interaction on a different angle. See manual 4.2.11 and figure 
referenced there.

> Suposed for an unkown molecule, how to assign them on rtp file?

Working by analogy with those of functional groups in existing .rtp 
definitions would be a good start. Find a similar moiety, write out the 
connectivity and improper definitions and see what is being done.

> Can you give some introduction or show somewhere that has the document?
> Gromacs menu does not document about this.

What documentation exists is probably in either the original CHARMM27 
force field files (download from web) or the papers that described the 
development (references in GROMACS manual). OPLS/AA apparently works 
differently, for some reason.

Mark

> Thanks for advance!
> Tom
>
>
>     Message: 1
>     Date: Sat, 11 Feb 2012 11:15:04 +1100
>     From: Mark Abraham <Mark.Abraham at anu.edu.au
>     <mailto:Mark.Abraham at anu.edu.au>>
>     Subject: Re: [gmx-users] improper dihedral angle in charmm and opls aa
>     To: Discussion list for GROMACS users <gmx-users at gromacs.org
>     <mailto:gmx-users at gromacs.org>>
>     Message-ID: <4F35B308.3080504 at anu.edu.au
>     <mailto:4F35B308.3080504 at anu.edu.au>>
>     Content-Type: text/plain; charset="iso-8859-1"
>
>     On 11/02/2012 9:54 AM, Tom wrote:
>     > Dear Gromacs Users
>     > I am confused about definition of improper dihedral angle in
>     charmm27,
>     > which is compared to oplsaa.
>     > Why charmm ff in gromacs give *double* items for the same
>     dihedral angle.
>     > e.g.  for charmm27 ASN
>     >         CG      ND2     CB      OD1
>     > CG      CB      ND2     OD1
>     >         ND2     CG      HD21    HD22
>     > ND2     CG      HD22    HD21
>
>     These are not duplicates. Atom ordering is significant. These
>     *improper*
>     dihedral angles are enforcing planarity of the amide moiety. This is
>     done by influencing dihedral angles along intra-atom vectors where
>     there
>     is no bond. OPLS/AA is working differently somehow.
>
>     Mark
>
>     > in opls aa for ASN
>     >     CG    CB    CA     C     dih_ASN_chi1_C_C_C_CO
>     >     CA    CB    CG   ND2     dih_ASN_chi2_C_C_CO_N
>     > Thanks for advance!
>     > Tom
>     > PS:
>     > the detail is as follows
>     > ------------------------------------------------
>     > *in charmm27.ff/aminoacids.rtp*
>     > [ ASN ]
>     >  [ atoms ]
>     >         N       NH1     -0.47   0
>     >         HN      H       0.31    1
>     >         CA      CT1     0.07    2
>     >         HA      HB      0.09    3
>     >         CB      CT2     -0.18   4
>     >         HB1     HA      0.09    5
>     >         HB2     HA      0.09    6
>     >         CG      CC      0.55    7
>     >         OD1     O       -0.55   8
>     >         ND2     NH2     -0.62   9
>     >         HD21    H       0.32    10
>     >         HD22    H       0.30    11
>     >         C       C       0.51    12
>     >         O       O       -0.51   13
>     >  [ bonds ]
>     >         CB      CA
>     >         CG      CB
>     >         ND2     CG
>     >         N       HN
>     >         N       CA
>     >         C       CA
>     >         C       +N
>     >         CA      HA
>     >         CB      HB1
>     >         CB      HB2
>     >         ND2     HD21
>     >         ND2     HD22
>     >         C       O
>     >         CG      OD1
>     >  [ impropers ]
>     >         N       -C      CA      HN
>     >         C       CA      +N      O
>     >         CG      ND2     CB      OD1
>     >         CG      CB      ND2     OD1
>     >         ND2     CG      HD21    HD22
>     >         ND2     CG      HD22    HD21
>     >  [ cmap ]
>     >         -C      N       CA      C       +N
>     > -------------------------------------------
>     > in*oplsaa.ff/aminoacids.rtp*
>     > [ ASN ]
>     >  [ atoms ]
>     >      N    opls_238   -0.500     0
>     >      H    opls_241    0.300     0
>     >     CA    opls_224B   0.140     1
>     >     HA    opls_140    0.060     1
>     >     CB    opls_136   -0.120     2
>     >    HB1    opls_140    0.060     2
>     >    HB2    opls_140    0.060     2
>     >     CG    opls_235    0.500     3
>     >    OD1    opls_236   -0.500     3
>     >    ND2    opls_237   -0.760     4
>     >   HD21    opls_240    0.380     4
>     >   HD22    opls_240    0.380     4
>     >      C    opls_235    0.500     5
>     >      O    opls_236   -0.500     5
>     >  [ bonds ]
>     >      N     H
>     >      N    CA
>     >     CA    HA
>     >     CA    CB
>     >     CA     C
>     >     CB   HB1
>     >     CB   HB2
>     >     CB    CG
>     >     CG   OD1
>     >     CG   ND2
>     >    ND2  HD21
>     >    ND2  HD22
>     >      C     O
>     >     -C     N
>     >  [ dihedrals ] ; override some with residue-specific ones
>     >      N    CA    CB    CG     dih_ASN_chi1_N_C_C_C
>     >     CG    CB    CA     C     dih_ASN_chi1_C_C_C_CO
>     >     CA    CB    CG   ND2     dih_ASN_chi2_C_C_CO_N
>     >
>     >
>
>     -------------- next part --------------
>     An HTML attachment was scrubbed...
>     URL:
>     http://lists.gromacs.org/pipermail/gmx-users/attachments/20120211/a4d6d18e/attachment-0001.html
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120212/73693af3/attachment.html>


More information about the gromacs.org_gmx-users mailing list