-
Notifications
You must be signed in to change notification settings - Fork 21
Syntax: build file
The build file (extension .bld) allows users to steer certain aspects of the structure generation and backmapping in polyply. At the moment polyply offers two types of options. Build options options are criteria that modify the acceptance of a step in the random-walk protocol in addition to the overlap criterion. For example, there are geometric restraints, which assess if a step within a region of space is valid or not. Another form of built option is a distance restraint that checks, if a step generates coordinates that obey specified distance restraints. Besides build options, there are two more options that specify more global aspects of the structure generation namely the volume of a residue and the template coordinates.
Polyply generates residue templates on the fly by a graph embedding procedure. However, the procedure becomes slow for large residues or it can fail to generate correct templates, especially if there are multiple chiral centers present. Even though a warning is emitted in those cases, it is more convenient to provide templates used in the structure generation. This is done by specifying a template using the [ template ] directive. Multiple templates can be given by providing the template directive multiple times. The syntax is shown below. Note that <> means replace keyword. Do not copy the < or > signs.
[ template ]
resname <resname>
[ atoms ]
<atomname> <atomtype> <x> <y> <z>
[ bonds ]
<atomname> <atomname>
The atomnames and atomtypes must exactly match those in the provided topology for the residue. Furthermore, there can only be one template per residue. Coordinates should be given in nm. If templates are provided, and no volumes (see next section) the volume is estimated from the residue template using the generic procedure.
As with the custom templates it sometimes can be convenient to provide template volumes to polyply instead of using the generic model. Volumes are defined as the sigma parameter of a Lennard-Jones interaction function. Cross interactions between residues at the moment are calculated from combination rules. The syntax is shown below. Note that <> means replace keyword. Do not copy the < or > signs.
[ volumes ]
<resname> <LJ_sigm_value>
Note that the unit of the LJ sigma value should be in nm. C6/C12 parameters are not supported in this input. If volumes are provided they override any volumes estimated.
Built options are grouped by molecule types and can be nested under each other. Therefore each entry starts with the [molecule] directive, which allows to select specific molecules from the topology to apply the build options to. The syntax of the molecule directive consists of the directive key and then on a new line the molecule name as listed in the topology and the molecule indexes as defined by order of appearance in the topology file. Note molecules start counting at 0. For example, to select the first 100 molecules in a PEO melt the molecule directive would be written as shown below:
[ molecule]
PEO 0 100
The build options are directives as well that can be listed under the '[molecule]' directive (see table below). Each directive consists again of the directive key and then on a new line a number of parameters describing the directives.
| build options | num. parms. | order of parameters and units | Comment |
|---|---|---|---|
| sphere | 8 | resname; resid_start; resid_stop; in/out; x (nm); y (nm); z (nm); r (nm) | resids define a range of resids |
| rectangle | 10 | resname; resid_start; resid_stop; in/out; x (nm); y (nm); z (nm); a (nm); b (nm); c (nm) | resids define a range of resids |
| cylinder | 9 | resname; resid_start; resid_stop; in/out; x (nm); y (nm); z (nm); r (nm); dz (nm) | resids define a range of resids |
| persistence_length | 4 | WCM; lp (nm); first resid ; last resid | resid define the first and last residue |
| distance_restraints | 3 | resid 1, resid 2, distance (nm) | resid define two different residues |
| rw_restriction | 4 | resname; resid_start; resid_stop; nx (nm); ny (nm); nz (nm); angle (deg) | n refers to normal |
For example, to place the first 10 polymer molecules with name polym and 100 residues of randomly placed PS, or PMA into a sphere of radius 4. located at a central point with coordinates x=10.0, y=10.0, z=10.0 the directive would be:
[ molecule ]
polym 0 10
[ sphere ]
PS 0 100 in 10.0 10.0 10.0 4.0
PMA 0 100 in 10.0 10.0 10.0 4.0
Note that here because we do not want to specify every single residue along the chain we simply set the resid range from 0 to 100. Polyply will print some warnings but correctly identify the residues we want.
Most directives require a residue name and a range of resids to be specified. This gives users the option to specifically target only certain residues within a molecule. Note that resids always start at counting 1 and is taken over the interval (start,stop]. Except for distance restraints and persistence length all residue specifications are only considered if the combination of molecule index, molecule name, residue name and resid all match. Otherwise they are skipped and a warning is issued. However, this allows very convenient writing of entries. For example, in a protein of 100 residues one wants to select only the ALA residues. Then using ALA 0 100 will result in only the Alanines being selected and all other residues being skipped instead of having to single out every Alanine residue.
Geometrical restraints check if a new point is within a certain volume in space and then either cause the algorithm to accept or reject the step. Regions in space can be defined as spheres, rectangles or cylinders, plus one of the keywords in or out to specify if the step needs to be inside or outside the volume. The regions in space are then defined by the following rules:
- a sphere is defined by it's center and a radius.
- a rectangle is defined by it's center and three distances from the center to the faces
- a cylinder is defined by it's center a radius, and the distance to the cylinder wall
Thus each directive consist of the x,y,z coordinates of the central point followed by the additional parameters defining the shape.
The persistence length directive contains the entries specifying how to sample the persistence length. It is important to be aware that the persistence length is taken into account by generating a end-to-end distance distribution and then sub-sampling that distribution to generate the end-to-end distances. However, when only a single polymer molecule is specified the algorithm will take one end-to-end distance from the distribution. In addition for very stiff molecules, if the polymer is too long the end-to-end distance is not a good descriptor anymore and instead the persistence length should be applied on segments of the chain.
The directive itself consists of one line. First the keyword WCM is given, which indicates that the worm-like-chain model is used. Currently only the WCM model is implemented. Following the WCM keyword the persistence length is given in nm. Afterwards two integer resids have to be given indicating the first and the last residue of the chain. Using this syntax also branched or grafted polymers can be treated by selecting the appropriate start and end point of the chain. Note that there has to be one unique shortest path connecting the two residues.
Note that you can add multiple persistence length directives for a molecule, for example, if you have a block-copolymer.
Distance restraints consist of two resid followed by a distance at which they need to be restrained. Note that distance restraints are by default satisfied up to +/- 1 step length. Multiple distance restraints and also overlapping restraints can be specified. However, for multiple restraints it does not mean the algorithm will find a solution. We perform no check as to whether the input is valid and can actually be satisfied. As always trying multiple times is a good cause of action in case the restraints cannot be satisfied.
The rw_restriction directive allows to restrict the direction of the random-walk along a specific vector. It works by specifying that vector and an angle. The vector describes a plane and the angle describes the maximum allowed angle of the new step with that plane. If the angle is 90 degrees it means the random-walk will not cross "back" across the plane and if it is even lower it means it will approx a step in direction of the plane normal. This option can be used to generate brushes as shown in the PEGylated lipid tutorial.