Force Field Explorer 

Home
Download
Getting Started
Documentation
Gallery
Source Code
Links
Acknowledgments

 

Getting Started

    The best way to get started using Force Field Explorer with TINKER is to try some of the example structures located in the "tinker/test" directory. These canonical systems have been used to address a wide range of thermodynamic and/or methodology questions by many groups in the field using a range of force fields. Keyword files are supplied and should provide an introduction to various options that can be specified to control execution of modeling commands. All images in the Gallery section of this web site are from systems found in the TINKER distribution.

Step 1: Open a TINKER coordinate file.

For example "tinker/test/enkephalin.xyz".

Step 2: Specify keywords that control aspects of the modeling commands.

     The default keywords for enkephalin, stored in the "tinker/test/enkephalin.key" file, were loaded automatically when the enkephalin coordinate file was opened. Using this set of keywords is a good place to begin. Notice that enkephalin uses the MM3PRO parameter set. Simply changing the "PARAMETER" keyword to a different force field parameter set does not change the atom type identifiers in the coordinate file.

Step 3: Execute a modeling command.

     The "NEWTON" minimizer is quite fast for smaller systems (where storage of the Hessian is not an issue) so it is a good choice for optimizing enkephalin to a local minimum. For protein size systems "MINIMIZE", which only uses first derivatives, is a safer choice.

Step 4: Observe structural and/or energetic results of the modeling command.

     The Newton optimization finds a local minimum for enkephalin after 60 iterations - less than a minute of CPU time on current hardware. TINKER creates a new structure file, "tinker/test/enkephalin.xyz_2", to store the local minimum. Force Field Explorer monitors the calculation, interactively updating coordinates, and loads enkephalin.xyz_2 over the top of enkephalin.xyz when TINKER completes.

Step 5: In a "real" molecular engineering setting, hopefully the investigator has learned something from the modeling command. They might then change a few keywords and/or execute another modeling command. In the future we hope to add editing features to close the design loop. In this case, try removing the implicit solvation model (set it to "absent") from the Keyword Editor and re-optimizing the structure.