Running a Go model with the AMHGMIN

GMIN can run various types of Go models with the AMH framework. This is useful for creating starting points for OPTIM, and also there are different forms of Go-like potentials that maybe interesting. To use this, first update and compile AMHGMIN as described in the GMIN Makefile.

Flags needed for GMIN input

The parameters behind the basin-hopping can be defined in the data file using the terms:

To begin with the AMH need to be turned on:

AMH

The interval of structures which are to be written to the file movie is termined by:

NINT_AMH    5

For structure prediction runs look something like this:

SLOPPYCONV  1.0D-3
TIGHTCONV   1.0D-3
UPDATES     5000
AMH
NINT_AMH    5
STEP  1.0 
MAXBFGS   1.0
TEMPERATURE 10.0

For slowly finding the lowest energy structure which is necessary as starting points for PATHSAMPLE runs shorter length steps, and much lower temperatures are necessary. The file then looks something like this:

SLOPPYCONV  1.0D-3
TIGHTCONV   1.0D-5
UPDATES     5000
AMH
NINT_AMH    5
STEP  0.4 
MAXBFGS   0.4
TEMPERATURE 1.0
MAXIT 1000 5000
RANSEED 1

Input files for AMHGMIN

There are several files that are necessary. Making sure the paths are correct is important. For running a Go model with AMHGMIN.

1. The primary input file is creatively called input_amh.

2. An alignment file which correlates the target sequence to the memory structures. In the most simple case there is one memory structure which is an experimental structure, but it does not need be. The following directory needs to be in the running directory. The directory in match/<filename>/filename. Below is an example of a match file. Where the name of the protein should be the file name and the name of the protein should be replaced in the first line of the file. The rest of the file is an identity alignment between the target sequence and the structure sequence for a protein 58 residues long.

1uama        1uama
   1   2   3   4   5   6   7   8   9  10
  11  12  13  14  15  16  17  18  19  20
  21  22  23  24  25  26  27  28  29  30
  31  32  33  34  35  36  37  38  39  40
  41  42  43  44  45  46  47  48  49  50
  51  52  53  54  55  56  57  58

3. The memory structure which defines the Go contacts is necessary. The following directory needs to be in the running directory. The directory in proteins/filename. The format of this file needs to be in the wolynes file format which contains the sequence, the coordinates, the secdondary structure definitions. and the surface accessibility. Converting the file can be done with pdb2wolynes.f program. This file is can be found in the svn/SCRIPTS/AMH directory. The file names must be 5 characters where the first 4 are the pdb ID and the final character is the chain within the pdb file.

4. The strength of these interactions is defined by gamma.dat roughly speaking the interaction energy should be equally divided between local, medium, and long sequence separations. The appropriate gamma.dat file is /home/mp466/amh/md_input/gammas/gamma.go.fourletter.

5. A directory of input parameters which add certain parameters for the backbone interactions. The directory to copy is /home/mp466/amh/params into the running directory params.

Moving these files around is defined in the shell scripts, so it does not take any human intervention.