Skip to content

FragIt FMO FD Example

Jump to bottom Edit New page
Casper Steinmann edited this page May 17, 2016 · 3 revisions

This example will help you setup an FMO frozen domain (FD) calculation using FragIt.

We start out by creating a configuration file

fragit-conf -o fmofd.ini

In FMO/FD, one has an active region A which is to be geometry optimized. Surrounding A is a buffer region b which has its electronic structure optimized using the same level of theory as A. A and b combined is called B. The rest of the structure is frozen and denoted F.

FMO/FD uses the multilayer formulation of FMO and the following layers are assigned the different regions:

  • F is assigned layer 1.
  • B is assigned layer 2. Remember that B includes A.

To control and setup an FMO/FD calculation using FragIt, the following four parameters, discussed in detail below, must be specified by the user:

  • The central fragment in the system.
  • The size of the active region.
  • The size of the buffer region.
  • The rough size of the layers.

We must first know which fragment is the central fragment of our system. This is specified in the [output] section in the configuration file with the variable centralfragment. Usually this is the fragment participating in a reaction or a chromophore. From this, all other sizes of different regions are determined by their minimum distance from that central fragment. When FragIt determines the size of different regions it is done for entire fragments, i.e. if one fragment has at least one atom within a certain distance from the central fragment, that entire fragment is included in a region (and not just that single atom).

First we must consider the size of the region B we wish to use (remember that B includes both A and b). For example we might choose this to be 7 Å. This is done by specifying boundaries = 7 in the [output] section of the configuration file. This sets an initial limit on the size of region B which now includes all fragments with at least one atom within 7 Å from the central fragment.

Next we consider the size of the active region A is determined by the user through the active keyword in the [output] section of the configuration file. For example we might choose this to be 5 Å so we specify active = 5.

To make an argument for the fourth option, buffer, we will discuss what we have so far. We have an active region, A, where all fragments within 5 Å are considered active and we have a region, B, which includes all fragments within 7 Å. The region b which is B but not A is called the buffering region or the link region. Because fragments are not uniformly sized or placed in a protein the size of b is thought to be 2 Å but this is not always the case. Since the FMO/FD formulation requires that A and F cannot be right next to each other, the buffer keyword extends b after A has been constructed to make sure that the distance between A and F is at least the value of buffer. This is done by promoting fragments from F with at least one atom within buffer distance from to b. Usually we choose buffer to have a value of boundaries minus active, i.e. 2 Å in this case, but this can be chosen freely by the user.

With these four options specified, run FragIt as your normally would

fragit --use-config=fmofd.ini structure.pdb
Add a custom footer
Add a custom sidebar
Clone this wiki locally