def test_sample_1():
# Test that the code actually runs and gives something non-crazy
# Make an ergodic dataset with two gaussian centers offset by 25 units.
chunk = np.random.normal(size=(20000, 3))
data = [np.vstack((chunk, chunk + 25)), np.vstack((chunk + 25, chunk))]
clusterer = cluster.KMeans(n_clusters=2)
msm = MarkovStateModel()
pipeline = sklearn.pipeline.Pipeline([("clusterer", clusterer),
("msm", msm)])
pipeline.fit(data)
trimmed_assignments = pipeline.transform(data)
# Now let's make make the output assignments start with
# zero at the first position.
i0 = trimmed_assignments[0][0]
if i0 == 1:
for m in trimmed_assignments:
m *= -1
m += 1
pairs = msm.draw_samples(trimmed_assignments, 2000)
samples = map_drawn_samples(pairs, data)
mu = np.mean(samples, axis=1)
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
# We should make sure we can sample from Trajectory objects too...
# Create a fake topology with 1 atom to match our input dataset
top = md.Topology.from_dataframe(pd.DataFrame({
"serial": [0],
"name": ["HN"],
"element": ["H"],
"resSeq": [1],
"resName": "RES",
"chainID": [0]
}),
bonds=np.zeros(shape=(0, 2), dtype='int'))
# np.newaxis reshapes the data to have a 40000 frames, 1 atom, 3 xyz
trajectories = [md.Trajectory(x[:, np.newaxis], top) for x in data]
trj_samples = map_drawn_samples(pairs, trajectories)
mu = np.array([t.xyz.mean(0)[0] for t in trj_samples])
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
def test_sample_1():
# Test that the code actually runs and gives something non-crazy
# Make an ergodic dataset with two gaussian centers offset by 25 units.
chunk = np.random.normal(size=(20000, 3))
data = [np.vstack((chunk, chunk + 25)), np.vstack((chunk + 25, chunk))]
clusterer = cluster.KMeans(n_clusters=2)
msm = MarkovStateModel()
pipeline = sklearn.pipeline.Pipeline(
[("clusterer", clusterer), ("msm", msm)]
)
pipeline.fit(data)
trimmed_assignments = pipeline.transform(data)
# Now let's make make the output assignments start with
# zero at the first position.
i0 = trimmed_assignments[0][0]
if i0 == 1:
for m in trimmed_assignments:
m *= -1
m += 1
pairs = msm.draw_samples(trimmed_assignments, 2000)
samples = map_drawn_samples(pairs, data)
mu = np.mean(samples, axis=1)
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
# We should make sure we can sample from Trajectory objects too...
# Create a fake topology with 1 atom to match our input dataset
top = md.Topology.from_dataframe(
pd.DataFrame({
"serial": [0], "name": ["HN"], "element": ["H"], "resSeq": [1],
"resName": "RES", "chainID": [0]
}), bonds=np.zeros(shape=(0, 2), dtype='int')
)
# np.newaxis reshapes the data to have a 40000 frames, 1 atom, 3 xyz
trajectories = [md.Trajectory(x[:, np.newaxis], top)
for x in data]
trj_samples = map_drawn_samples(pairs, trajectories)
mu = np.array([t.xyz.mean(0)[0] for t in trj_samples])
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
n_first = 17
n_clusters = 9
slicer = featurizer.FirstSlicer(n_first)
clusterer = cluster.KMeans(n_clusters=n_clusters)
msm_model = msm.MarkovStateModel()
pipeline = make_pipeline(slicer, clusterer, msm_model)
s = pipeline.fit_transform(X)
p0 = make_pipeline(dih_model, tica_model, slicer)
trajectories = dataset.MDTrajDataset("./trajectories/*.h5")
selected_pairs_by_state = msm_model.draw_samples(s, 5)
samples = utils.map_drawn_samples(selected_pairs_by_state, trajectories)
for k, t in enumerate(samples):
t.save("./pdbs/state%d.pdb" % k)
Y = p0.transform(samples)
hexbin(Xf[:, 0], Xf[:, 1], bins='log')
plot(clusterer.cluster_centers_[:, 0],
clusterer.cluster_centers_[:, 1],
'k+',
markersize=12,
markeredgewidth=3)
map(lambda k: annotate(k, xy=clusterer.cluster_centers_[k, 0:2], fontsize=24),
arange(n_clusters))
#map(lambda y: plot(y[:, 0], y[:, 1], 'x', markersize=8, markeredgewidth=2), Y)
src_samples = msm_src.draw_samples(raw_dtrajs_abl,n_samples_per_state)
# # coarse-grain abl?
# from msmbuilder import lumping
# n_macrostates = 16
# pcca_abl = lumping.PCCAPlus.from_msm(msm_abl, n_macrostates=n_macrostates)
# abl_samples = pcca.draw_samples(dtrajs_abl,n_samples_per_state)
#
# # coarse-grain src?
# n_macrostates = 24
# pcca = lumping.PCCAPlus.from_msm(msm_src, n_macrostates=n_macrostates)
# src_samples = pcca.draw_samples(dtrajs_abl,n_samples_per_state)
# fetch corresponding configurations
from glob import glob
traj_files_abl = glob('../abl_snapshot/*.h5')
traj_files_src = glob('../src_snapshot/*.h5')
from msmbuilder.utils import map_drawn_samples
frames_abl = map_drawn_samples(abl_samples,traj_files_abl)
frames_src = map_drawn_samples(src_samples,traj_files_src)
# to do : wrap in a function
for i in range(len(frames_abl)):
msm_weight = msm_abl.populations_[i]
frames_abl[i].save_pdb('abl_cluster_{0}_msm_weight_{1:.3f}.pdb'.format(i,msm_weight))
for i in range(len(frames_src)):
msm_weight = msm_src.populations_[i]
frames_src[i].save_pdb('src_cluster_{0}_msm_weight_{1:.3f}.pdb'.format(i,msm_weight))
tica_model = utils.load("./tica.pkl")
dih_model = utils.load("./dihedrals/model.pkl")
n_first = 17
n_clusters = 9
slicer = featurizer.FirstSlicer(n_first)
clusterer = cluster.KMeans(n_clusters=n_clusters)
msm_model = msm.MarkovStateModel()
pipeline = make_pipeline(slicer, clusterer, msm_model)
s = pipeline.fit_transform(X)
p0 = make_pipeline(dih_model, tica_model, slicer)
trajectories = dataset.MDTrajDataset("./trajectories/*.h5")
selected_pairs_by_state = msm_model.draw_samples(s, 5)
samples = utils.map_drawn_samples(selected_pairs_by_state, trajectories)
for k, t in enumerate(samples):
t.save("./pdbs/state%d.pdb" % k)
Y = p0.transform(samples)
hexbin(Xf[:, 0], Xf[:, 1], bins='log')
plot(clusterer.cluster_centers_[:, 0], clusterer.cluster_centers_[:, 1], 'k+', markersize=12, markeredgewidth=3)
map(lambda k: annotate(k, xy=clusterer.cluster_centers_[k, 0:2], fontsize=24), arange(n_clusters))
#map(lambda y: plot(y[:, 0], y[:, 1], 'x', markersize=8, markeredgewidth=2), Y)