Non-aqueous solvent

---------------------------- 4.1 and later versions:

The easiest approach is to use the new model builder LEaP, which supports different solvent models. Or see the new `GEN' option in EDIT.

---------------------------- 4.0 and previous versions:

The following text are responses to this basic question posted to the Computation Chemistry and/or AMBER Mail Reflector. The test been HTML-ized somewhat...

Some types of non-aqueous simulations can be done with the AMBER code as is. I have run some periodic boundary simulations of a peptide in methanol. The steps involved are:

1. add the methanol parameters to the database. I used the parameters from Jorgensen:
   J. Phys. Chem. 94, 1683 (1990)
   J. Phys. Chem. 90, 1276 (1986).
2. Create a PDB file with your solute surrounded by the appropriate box of solvent molecules.
3. Create a prep file for you solvent molecule.
4. Create a link.input file defining the solute and surrounding solvent molecules. Define each solvent molecule as a separate molecule in link.
5. Create an edit.input file that creates a "dummy" box. This means you use the "BOX" option to add waters, but set the closeness of approach parameters so that no waters will actually be added. You should also use the "XRAY" option to read in the PDB file you created above.
6. Create a parm file for the system. Now edit the resulting topology file. Do the following:
   • Search the file for a line starting with either "0.90000000E+02" or " .90000000E+02" (depending on whether your system writes the leading 0). This line contains the dimensions of the box. You will need to know these, and replace them in this file. The values on this line are:
     

                angle        boxx    boxy    boxz
     	    

     Change only the box dimensions to reflect the size of the box in your PDB file.
   • Search *backwards* in the file, starting with the second line above the line you modified above, for the first line that contains three, and only three, values. The values on this line are:
     

                IPTRES       NSPM    NSPSOL
     	    

     IPTRES: Set to the number of the last *residue* to be considered part of the solute.
     NSPM: Do not modify.
     NSPSOL: Set to the number of the first *molecule* to be considered part of the solvent.
     
7. If you are running with SANDER/MINMD, you're done. If you are running with GIBBS, you should set the variable ITIP in the input to 1, to indicate you are using a non-water solvent.

You might consider directing any further queries regarding AMBER to the mailing list maintained at UCSF. To subscribe to the list, just send mail to amber-request@cgl.ucsf.edu. To send queries, comments, whatever, just send mail to amber@cgl.ucsf.edu. I have had good responses from people who read that list.

  NOTE: the following address may be out of date! (tec3)
> Good Luck!


Charles K. Foley, Ph.D.
Cray Research, Inc.
Americas & Asia/Pacific Group
P.O. Box 12746
Research Triangle Park, NC 27709                                o
                                                              <(
ckf@f16.cray.com                                               [\.

Sorry, not an Amber user (yet). But for GROMOS, there should be MeOH, CCl4 (but there's a paper, telling full MD in CCl4 is not much better than SD with appropriate parameters) and DMSO (Rao+Singh: JACS 112, 3803-3811 (1990); Mierke+Kessler: JACS 113, 9466-9470 (1991) + one more recently, I think). I seem to recall some rumours about CHCl3, but I might be veeery wrong ...

+======================================+=====================================+
|  Wolfgang Sauer                      |                                     |
|  Institut fuer Organische Chemie I   |  "You have to be three standard     |
|  Henkestr. 42, D-8520  Erlangen, FRG |   deviations away from a normal     |
|                                      |       personality to like UNIX."    |
|  sauer@organik.uni-erlangen.de       |                                     |
|  Tel.: 49/0 - 9131 - 85 - 2952       |              Morris Jones (C&T)     |
|  Fax:                   - 9132       |                                     |
+======================================+=====================================+