added gromos rmsd clustering

[jamesmkrieger]

This method is the same as the one used in https://manual.gromacs.org/documentation/current/onlinehelp/gmx-cluster.html

[jamesmkrieger]

Perhaps we should give this a better name

[jamesmkrieger]

Definitely, something isn't right here for the labels=None case. Perhaps make labels a necessary argument rather than defaulting to None

Actually, we do want to support the case where we have no labels, e.g. from an MD trajectory. It was just done wrong.

[jamesmkrieger]

It should be good now, but needs testing

[jamesmkrieger]

This works to cluster a set of trajectories now:

[jamesmkrieger]

from prody import *
import matplotlib.pyplot as plt
plt.ion()
import numpy as np

ref_pca = loadModel('prody_pca/ref.pca.npz')

ens = parseDCD("combined.dcd", step=100)
ens.iterpose(rmsd=0.01)
rmsd_matrix = ens.getRMSDs(pairwise=True)

x = calcRMSDclusters(rmsd_matrix, 2.1)

colors = np.zeros(num_confs, dtype=int)
for i, c in enumerate(x):
    colors[list(np.array(c, dtype=int))] = i
colors = list(colors)

plt.figure()
showProjection(ens, ref_pca[:3], markersize=3, color=colors)
plt.savefig('gromos_clusters_2p1A.png')