Water biasing

One feature associated with the latest version of the Go model is the ability to bias the non-bonded interactions with water, specified by secondary structure. As the reference demonstrates, this may be important in fixing several problems with the current model of proteins, including over-compactness of intrinsically disordered regions.

The documentation describes these features:

Apply water bias.:
  -water-bias           Automatically apply water bias to different secondary structure elements. (default: False)
  -water-bias-eps WATER_BIAS_EPS [WATER_BIAS_EPS ...]
                        Define the strength of the water bias by secondary structure type. For example, use `H:3.6 C:2.1` to bias helixes and coils. Using
                        the idr option (e.g. idr:2.1) intrinsically disordered regions are biased seperately. (default: [])
  -id-regions WATER_IDRS [WATER_IDRS ...]
                        Intrinsically disordered regions specified by resid.These parts are biased differently when applying a water bias.format:
                        <chain>-<start_resid_1>:<end_resid_1> <chain>-<start_resid_2>:<end_resid_2>... (default: [])
  -idr-tune             Tune the idr regions with specific bonded potentials. (default: False)

These flags can be specified in conjunction with the Go model.

Water biasing for secondary structure

To apply a water bias to your protein dependent on the secondary structure, the first two flags described above must be used.

martinize2 -f protein.pdb -o topol.top -x cg_protein.pdb -dssp -water-bias -water-bias-eps H:1

This will produce a coarse-grained model of your protein, with virtual sites along the backbone. The virtual sites will be defined in an external file, which should be included in your topology as per the Go model instructions.

There will also be a second file, defining the additional non-bonded interactions between water and the secondary structure elements defined in the command. In this case, any residue identified as H (i.e. helix) by dssp will have an additional Lennard-Jones interaction of epsilon = 1 kJ/mol between its backbone virtual site and water.

To define more interactions based on secondary structure, add more letter codes to the -water-bias-eps:

martinize2 -f protein.pdb -o topol.top -x cg_protein.pdb -dssp -water-bias -water-bias-eps H:1 C:0.5 E:2

Water biasing for intrinsically disordered regions/proteins

If you have disordered regions in your protein, then they can have additional bonded and nonbonded parameters added (described more in the Go model paper).

These regions need to firstly be annotated by the user, using the -id-regions flag to indicate resid segments known to be disordered:

martinize2 -f protein.pdb -o topol.top -x cg_protein.pdb -dssp -id-regions A-1:10 B-65:92

Ideally, as the paper describes, these should have their water bias and bonded parameters fixed too. This can be done by combining the above command with the ones previously described about water biasing:

martinize2 -f protein.pdb -o topol.top -x cg_protein.pdb -dssp -id-regions A-1:10 B-65:92 -idr-tune -water-bias -water-bias-eps idr:0.5

Here, -idr-tune makes sure that the additional bonded parameters are applied to the region specified by -id-regions, while -water-bias and -water-bias-eps idr:0.5 ensures that for the idr region defined, an additional nonbonded parameter with water is written to the nonbond_params.itp file.

For a single chain, or a homomultimer containing identical disordered regions, the chain specifier on the -id-regions flag is not necessary. The command:

martinize2 -f protein.pdb -o topol.top -x cg_protein.pdb -dssp -id-regions 50:75 -idr-tune -water-bias -water-bias-eps idr:0.5

will apply disordered parameters and biasing to residues 50:75 of all chains in the system.

If you’re working extensively with proteins which are fully disordered in Martini, it may be more convenient to use Polyply to generate the input parameters for your system than Martinize2, as Polyply does not require an atomistic input structure to generate these parameters. The tutorial on the Polyply wiki may be a useful starting point as an indication for Polyply can be used for this.