[gmx-users] add a group to an amino acid

[Mark Abraham]

On 14/11/2010 7:26 PM, hengame fallah wrote:
> Dear Mark,
> I think that i couldn't explain my problem clearly, however i have 
> changed my .rtp but i don't know how should be my .hdb file.
>
> a) The structure of *_BOC_* is: *-CO-CH2-CH(C6H5Cl)-CH2-NH-*

Great.

> b) This is the new charges:
> [ BOC ]
>  [ atoms ]
>      N    opls_238   -0.500       1
>      H    opls_241    0.300       1
>     CA    opls_224B   0.080     1
>    HA1    opls_140    0.060     1
>    HA2    opls_140    0.060     1
>     CB    opls_149    0.055      2
>     HB    opls_140    0.060      2
>    CG1    opls_145   -0.115     2
>    CD1    opls_145   -0.115     3
>    HD1    opls_146    0.115     3
>    CD2    opls_145   -0.115     4
>    HD2    opls_146    0.115     4
>    CE1    opls_145   -0.115     5
>    HE1    opls_146    0.115     5
>    CE2    opls_145   -0.115     6
>    HE2    opls_146    0.115     6
>     CZ    opls_145    1.000      7
>     Cl    opls_264   -1.000       7
>    CG2    opls_071   -0.120     8
>    HG1    opls_140    0.060     8
>    HG2    opls_140    0.060     8
>      C    opls_235    0.700       9
>      O    opls_236   -0.700       9

This looks much better.

> c) charge for CG1 in PHE (in OPLS .rtp) is negative, too.

OK

> d) Would you explain for me that what is wrong in my atom types?

No. My point last email was that you may need to consider what atom 
types are suitable for your backbone methylene C. I don't know what is 
suitable. You will have to look at the .atp file and perhaps read 
OPLS/AA papers.

> e) As you see in the error, at first there are 63 residues (like my 
> pdb) but when when it adds COO- and NH3+ there are 69 residues! it 
> should add only one "O" (for COO-) and two "H"s for (NH3+) i think.

Well, what do the .tdb files say? You should model these for BOC upon 
the PHE ones, of course. You will need to have read the parts of chapter 
5 that pertain to the .hdb and .tdb files.

>
> There are 1 chains and 0 blocks of water and 3 residues with 63 atoms
>
>   chain  #res #atoms
>   1 ' '     3     63
>
> All occupancy fields zero. This is probably not an X-Ray structure
> Opening library file /usr/share/gromacs/top/ffoplsaa.atp
> Atomtype 1
> Reading residue database... (ffoplsaa)
> Opening library file /usr/share/gromacs/top/ffoplsaa.rtp
> Residue 57
> Sorting it all out...
> Opening library file /usr/share/gromacs/top/ffoplsaa.hdb
> Opening library file /usr/share/gromacs/top/ffoplsaa-n.tdb
> Opening library file /usr/share/gromacs/top/ffoplsaa-c.tdb
>
> Back Off! I just backed up topol.top to ./#topol.top.3#
> Processing chain 1 (63 atoms, 3 residues)
> There are 3 donors and 3 acceptors
> There are 4 hydrogen bonds
> Checking for duplicate atoms....
> Opening library file /usr/share/gromacs/top/specbond.dat
> 7 out of 7 lines of specbond.dat converted succesfully
> N-terminus: NH3+
> C-terminus: COO-
> Now there are 3 residues with 69 atoms
> Making bonds...
> Warning: Long Bond (53-54 = 0.334847 nm)
> Warning: Long Bond (53-56 = 0.334912 nm)
> Warning: Long Bond (64-65 = 0.271173 nm)
> Warning: Long Bond (64-66 = 0.347808 nm)
> Warning: Long Bond (64-67 = 0.31288 nm)

These warnings are still suggestive that your atom coordinates are 
labelled inappropriately in your coordinate file.

> Opening library file /usr/share/gromacs/top/aminoacids.dat
>
> -------------------------------------------------------
> Program pdb2gmx, VERSION 4.0.7
> Source code file: ../../../../src/kernel/add_par.c, line: 233
>
> Fatal error:
> Atom HB1 not found in rtp database in residue BOC, it looks a bit like HB

I think your .hdb is still trying to add two hydrogens to CB.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20101114/66a3fb9b/attachment.html>