[gmx-users] Tyrosine vsite aromatics

[David Chan]

Hi everyone

I am having some difficulty using 'vsite aromatics' for tyrosine 
residues. I've made a pdb of a sole tyrosine residue and run it through:

pdb2gmx -f tyr.pdb -o tyr.gro -vsite aromatics -heavyh

using gromacs 3.3 and ffG43a2

Subsequently I put it into a box (editconf) and try to run an EM at 
which point it explodes. ("This usually indicates your system is 
exploding." msg in EM) When I run the same procedure with -vsite 
hydrogens, the problem disappears.

I'm a bit confused because the gromacs 3.3 manual (e.g. Figure 6.4) 
shows which dummy atoms should be generated and indicates that HH should 
NOT be a virtual site. However, when I look in my topol.top file, the HH 
atom is treated as a virtual site with no mass. In addition, the program 
creates a virtual atom (MW1) in line with the OH-HH vector, 2 A from the 
HH atom. As far as I can tell from the manual that virtual site should 
not be there (?).

[ atoms ]
;   nr       type  resnr residue  atom   cgnr     charge       mass  
typeB    chargeB      massB
    1       MNH3      1    TYR    MN1      1          0    8.51535   ; 
qtot 0
    2       MNH3      1    TYR    MN2      1          0    8.51535   ; 
qtot 0
    3         NL      1    TYR      N      1      0.129          0   ; 
qtot 0.129
    4          H      1    TYR     H1      1      0.248          0   ; 
qtot 0.377
    5          H      1    TYR     H2      1      0.248          0   ; 
qtot 0.625
    6          H      1    TYR     H3      1      0.248          0   ; 
qtot 0.873
    7        CH1      1    TYR     CA      1      0.127     13.019   ; 
qtot 1
    8        CH2      1    TYR     CB      2          0     14.027   ; 
qtot 1
    9          C      1    TYR     CG      2          0     25.366   ; 
qtot 1
   10          C      1    TYR    CD1      3       -0.1          0   ; 
qtot 0.9
   11         HC      1    TYR    HD1      3        0.1          0   ; 
qtot 1
   12          C      1    TYR    CD2      4       -0.1          0   ; 
qtot 0.9
   13         HC      1    TYR    HD2      4        0.1          0   ; 
qtot 1
   14          C      1    TYR    CE1      5       -0.1     25.366   ; 
qtot 0.9
   15         HC      1    TYR    HE1      5        0.1          0   ; 
qtot 1
   16          C      1    TYR    CE2      6       -0.1     25.366   ; 
qtot 0.9
   17         HC      1    TYR    HE2      6        0.1          0   ; 
qtot 1
   18          C      1    TYR     CZ      7       0.15          0   ; 
qtot 1.15
   19         OA      1    TYR     OH      7     -0.548    16.5034   ; 
qtot 0.602
   20         MW      1    TYR    MW1      7          0      0.504   ; 
qtot 0.602
   21          H      1    TYR     HH      8      0.398          0   ; 
qtot 1
   22          C      1    TYR      C      9       0.27     12.011   ; 
qtot 1.27
   23         OM      1    TYR     O1      9     -0.635    15.9994   ; 
qtot 0.635
   24         OM      1    TYR     O2      9     -0.635    15.9994   ; 
qtot 0

I searched the mailing list and was only able to find this thread that 
seems to be related, but not quite the same: 
http://www.gromacs.org/pipermail/gmx-developers/2002-June/000239.html

This seems odd to me as this should have created problems earlier, no? 
Or am I missing something obvious?!

Thanks a lot for any suggestions!

David