Modules for modeling of molecular structure

Module for class ‘Structure’

There is a module, that realizes data structure - molecule and methods for its modification.

class kvazar.core.structure.Structure(struct_data, auto_calculate=True)

• atom_names - Array of names of atoms, i.e. ‘Carbon’, ‘Hydrogen’ and etc.
• atom_types - Numbers, that are correspondind to atoms names
• n_atoms - Number of atoms in structure
• n_links - Number of maximum possible links for atom. The second dimension of links matrix
• coords - Array of coordinates of each atom in structure
• velocity - Array of atoms velocities in form
• tension - Array of tensions for all atoms
• charges - Array of electric charge on all atoms
• links - Array of links of each atom with all others (one-sided links)
• num_links - Number of links for each atom
• full_links - Array of links for all atoms (two-sided bonds)
• num_full_links - Array of numbers of two-sided links for each atom

compare_topology_with(b)

It only compares number of atoms, links and identity of atom types in two structures.

compare_with(b)

If the structures are the same, it returns True.

copy()

Returns new structure which is equal to the initial

get_full_links(links)

It makes 2D array of reciprocal links between atoms.

get_neighbours(full_links)

It returns array of neighbours of all atoms, array of numbers of all atoms, full number of atoms

get_tors_sequences(full_links)

It finds sequences of atoms, that form dihedral angles

getsubstr(selected)

It forms new structure from selected atoms. For example:

Some_structure.getsubstr([(0,20),(30,45)]) - it returns new object of class Structure, that is made from 0-20 and 30-45 atoms of initial structure.

tors_dfs(full_links, pos, used, cur_way, result)

Depth-first search to find dihedral angles

update_neighbours()

It updates neighbours matrix, matrix of number of neighbours

There are some overrided methods for structure, as __add__, __iadd__, __getslice__, __delitem__.

And here is a python code of example of usage of module:

struct_data = dict()
n_atoms = 2
n_links = 10
struct_data['n_atoms'] = n_atoms
struct_data['n_links'] = n_links
struct_data['atom_names'] = ['Carbon','Carbon']
struct_data['coords'] = numpy.zeros((n_atoms,3), numpy.float)
struct_data['coords'][1][0] = 1.42
struct_data['links'] = numpy.zeros((n_atoms,n_links), numpy.int32)-1
struct_data['links'][0][0] = 1
struct_data['num_links'] = numpy.zeros(n_atoms, numpy.float)
struct_data['num_links'][0] = 1
A = structure.Structure(struct_data)
B = A.copy()
B.coords[0][1] = B.coords[1][1] = 1.42
C = A + B

You will get new structure with 4 atoms, where coordinates are concatenated, when velocities, links are still the same as in original A and B

A += B

You will get modified structure A (A = A + B)

D = A[1:2]

You will get new structure, which is a part of structure A and has 2 atoms (from 1 to 2 inclusive).

del A[[(0,0), (1,2)]]

It deletes atoms of structure: 0,1,2. It changes initial structure.

E = C.getsubstr([(1,2)])

You will get new structure, that made of atoms 1,2.

Module of class for structure loading and saving

class kvazar.fileman.FileManager

It’s a class for structure loading from files. Possible extensions of files for loading are ACD, ACC, PDB. Now there is only one possible extension for saving: ACD. Also class checks existance of direcory and makes the new one, if needed. Raise exception in the case of unknown extension.

There is an example of fileman.py usage in package:

from kvazar.fileman import FileManager
fm = FileManager()
struct1 = fm.load("path/to/c28.ACD")

It loads model of fullerene C28

fm.set_struct(struct1)
fm.save("path/to/c28_new.ACD")

It saves copy of model of fullerene in new file. Examples of input files can be found in kvazar’s directory: Path_to_kvazar + “/test/test_data/”

Module for saving output file

class kvazar.record.Record(rec_file, rec_type='w')

It’s a class for saving intermediate data of calculation and archive it in one output file.

It provides an interface to the molecular dynamic data. It allows:

• To load any state of calculated file. States are from 0 to the last of states

python
from kvazar.record import Record
rec = Record('path/to/file.nmd', 'a')
rec.update_state(1)

• To see parameters, that were calculated, for any state:

rec.params

There are two possible modifiers in initializing of object of class Record:

• ‘a’ - it will append new data to already existed
• ‘w’ - it will rewrite data