ClustIce

A tool to generate exotic ice clusters.

version 0.7.1

Requirements

• python
• numpy
• networkx
• scikit-learn
• cycless

Installation

ClustIce is registered to PyPI (Python Package Index). Install with pip3.

pip3 install clustice

Uninstallation

pip3 uninstall clustice

API

API manual is here.

Examples

Make an icosahedral ice cluster.

import numpy as np
import networkx as nx

import genice_core
from clustice.geometry import make_layout
from clustice.gromacs import render
from clustice.water import tip4p
from clustice import graph

# O-O distance
L = 0.27

# note: g must be a graph whose labels start from 0.
# g = graph.great_icosahedron(12, separation=L)
g = graph.great_decahedron(12)
# g = graph.small_barrelan()
# g = graph.large_barrelan()
# g = graph.twistane()
# g = graph.adamantane()
# g = nx.cycle_graph(6) # hexagon
# g = nx.cycle_graph(7) # heptagon
# g = nx.cubical_graph() # cubic octamer
# g = nx.dodecahedral_graph()


if "pos" in g.nodes[0]:
    # extract the embedded coords in g
    layout = np.array([g.nodes[v]["pos"] for v in g])
else:
    # estimate of the positions of the nodes
    layout = make_layout(g, edgelen=L)

# set orientations of the hydrogen bonds.
# if vertexPositions is given, the net dipole moment is minimized.
dg = genice_core.ice_graph(g, vertexPositions=layout, dipoleOptimizationCycles=100)

# put water molecules
gro = render(dg, layout, watermodel=tip4p)
with open("sample.gro", "w") as f:
    f.write(gro)