Personal code for calculating the solvent-accessible surface area of a molecule from GROMACS simulation output using a Monte Carlo technique. Random points are generated around each site located at a distance equal to the sum of the solvent probe radius and VDW radius for that site. A generated point is rejected if it is closer to another atomic site in the molecule than the site it was generated around. The ratio of points accepted to the number of tries gives the fraction of the surface area of a sphere for each atomic site. The summation of the solvent-accessible surface area of the atomic sites is the total SASA of the molecule.
# make PREFIX=/usr/local install
$ sasa config.json
Configuration files use the JSON format. The following options are available:
input.xtcfile- The trajectory file from your simulation.
input.ndxfile- The index file corresponding with the xtcfile, containing *ndxgroup`.
input.ndxgroup- The index group indicating the solute molecule.
config.r- Radius (nm) of solvent probe. The default is 0.14 nm.
config.rvdw- Array of VDW radii for the solute, which will be added to the solvent probe radius in the calculation. These need to be in the exact order of the sites listed in ndxgrp. Alternatively, only specify one VDW radius and it will be used for all sites. If no VDW radii are specified, 0.2 nm is used for each site.
config.nrand- Number of attempts for inserting a point around each site. The default is 1000.
config.nblocks- Number of blocks used in finding the blocked standard deviation, which is output as the uncertainty. The default is 5.
output.file- Where the output is going to be saved. The default is "sasa.txt". Results are in nm².
An example configuration file and other input files is in the
To run the example after compilation do:
$ ./bin/sasa example/config.json
The output file will be located at