[gmx-users] pdb2gmx problems and modified AA's

[Dr Seth Carlton OLSEN]

Hi Gromacs Users,

I am trying to run a pdb through pdb2gmx to get files to do an md run.  The protein chain contains an unconventional residue ('CRQ', in the pdb).  I am using the OPLS-AA forcefield.  I have modified the ffoplsaa.rtp, ffoplsaa.hdb and aminoacids.dat files by adding the following sections:
**added to ffoplsaa.rtp
;CRQ - chromophore GYQ from Rtms5 structure
[CRQ]
 [ atoms ]
    CD1    opls_145    0.029     1
   HD11    opls_146    0.000     1
    CE1    opls_145    0.030     1
   HE11    opls_146    0.000     1
    CZ     opls_166    0.204     1
    OH     opls_167   -1.316     1
    CE2    opls_145    0.029     1
   HE21    opls_146    0.000     1
    CD2    opls_145    0.029     1
   HD21    opls_146    0.000     1
    CG2    opls_145    0.050     1
    CB2    opls_142    0.064     1
   HB21    opls_144    0.000     1
    CA2    opls_141    0.204     1
    N2     opls_559   -0.436     1
    C2     opls_235    0.340     1
    O2     opls_236   -0.632     1
    N3     opls_557    0.110     1
    CA3    opls_223    0.134     1
   HA31    opls_140    0.000     1
   HA32    opls_140    0.000     1
    C      opls_235    0.433     1
    O      opls_236   -1.316     1
    C1     opls_558    0.340     1
    CA1    opls_257    0.204     1
    N      opls_251   -0.838     1
    CB1    opls_136    0.133     1
   HB11    opls_140    0.000     1
   HB12    opls_140    0.000     1
    CG1    opls_136    0.133     1
   HG11    opls_140    0.000     1
   HG12    opls_140    0.000     1
    CD3    opls_139    0.433     1
    OE1    opls_235   -0.632     1
    NE1    opls_237    0.127     1
    HAE    opls_909    0.072     1
    HAD    opls_909    0.072     1
 [ bonds ]			         
   CD1   CE1
   CD1  HD11
   CD1   CG2
   CE1    CZ
   CE1  HE11
    CZ    OH
    CZ   CE2
   CE2   CD2
   CE2  HE21
   CD2   CG2
   CD2  HD21
   CG2   CB2
   CB2   CA2
   CB2  HB21
   CA2    N2
   CA2    C2
    N2    C1
    C2    O2
    C2    N3
    N3   CA3
    N3    C1
   CA3     C
   CA3  HA31
   CA3  HA32
     C     O
    C1   CA1
   CA1     N
   CA1   CB1
   CB1   CG1
   CB1  HB11
   CB1  HB12
   CG1   CD3
   CG1  HG11
   CG1  HG12
   CD3   OE1
   CD3   NE1
   NE1   HAE
   NE1   HAD
    -C     N
[ impropers ]
    CA    +N     C     O    improper_O_C_X_Y

***added to ffoplsaa.hdb
CRQ     9
	1	1	CD1	CE1	CG2
	1	1	CE1	OH	CD1
	1	1	CD2	CE2	CG2
	1	1	CE2	OH	CD2
	1	1	CB2	CG2	CA2
	2	6	CB1	CA1	CG1
	2	6	CG1	CB1	CD3
	2	3	NE1	CD3	OE1
	2	6	CA3	N3	C

***added to aminoacids.dat
CRQ


when I run pdb2gmx on the pdb file 'pdb1mou.ent' (too large to be included here) I get the following error, which kills pdb2gmx:
Fatal error: Atom HE12 not found in rtp database in residue CRQ, it looks a bit like HE11

However, I am not sure why pdb2gmx is looking for this atom.  HE12 is not included in the pdb file (which, as expected, does not contain hydrogens).  Why is pdb2gmx looking for an atom that does not exist and should not?  Is there a naming convention in operation of which I am not aware?  Is pdb2gmx anticipating the naming of the hydrogens to be put in the structure?  

If anyone has any insight into this problem, I would  be very grateful.  For the moment, it has stopped me in my tracks.

Cheers,

Seth



ccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccms

Dr Seth Olsen, PhD
Postdoctoral Fellow, Computational Systems Biology Group
Centre for Computational Molecular Science
Chemistry Building,
The University of Queensland
Qld 4072, Brisbane, Australia

tel (617) 33653732
fax (617) 33654623
email: s.olsen1 at uq.edu.au
Web: www.ccms.uq.edu.au 

ccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccms