Current Polymer Builder
The polymer builder functionality currently in Towhee (as of version 7.1.0) is focussed on several popular special case polymers.
They each have their own specific option for building molecule connectivity maps associated with different values for the input_style
variable in towhee_input.
The nanotube builder in an elegant tool that allows easy construction of both the molecular parameters for a carbon nanotube, and the assignment of initial coordinates of the nanotube.
It is an excellent example of a specialized tool that performs a single task, but performs it very well. If you want to simulate a nanotube you just have to assign the appropriate
dimensions and select a force field and you are ready to begin.
In contrast, the polypeptide builder and the nucleic acid builder are the front end for a series of subroutines that are each specialized for a single force field and polymer combination.
They are designed to construct a specific set of heteropolymers that are commonly seen in biological systems. The specialized subroutines that perform these tasks are named poly*.F, where
the * is replaced with the name of a specific force field that contains the mapping required to piece together the relevant heteropolymers. There has been some effort to reduce duplication
by placing general templates into the monomers.F subroutine. Also in contrast with the nanotube builder, users typically have an initial structure (often a pdb file) that they wish to use
as the starting structure for the heteropolymer. While it might naively seem a straight-forward task to turn the pdb file coordinates into the initial structure
(using the pdb2towhee utility) , there is almost always
some cleanup and atom reordering required in order to deal with missing atoms in the pdb file. Even for cases where the pdb file is complete, often the atoms are listed in a slightly different
order between the Towhee assembly process and the pdb file, this results in mismatched coordinates for the initial structure and the common indicator of this problem is a warning message about
excessivly long bond lengths. Typically, the only way to resolve those issues is to compare the atom names in the towhee_altinp output file with
the atom names in the towhee_coords or towhee_partial file you translated from your pdb file.
Proposed Polymer Builder
While the carbon nanotube builder is fine as a specialized tool, all of the other polypeptide and polynucleic acid builders should be combined into a single utility that
assembles monomer files into any heteropolymer. This could be accomplished following the same path as the forcefield utility. Much like the polymer builder, all of the
various force field assignement subroutine were once built into the main Towhee code. These should be split out from the main code and used to generate formatted text files that contain the momomer
information in the same manner as the formatted text files that contain the force field information. Done properly, this would also allow users to create their own monomer text files to either
contribute to the code base, or for use to build their own polymers. I expect it would take a month of work to implement and document this properly. If anyone is interested in either performing
this implementation, or has ideas for funding such a project, please feel free to contact me about it.
Return to the Towhee algorithm page