def fit_bootstrap(yaml_file,pool=None):
mdl_params = yaml_file["mdl_params"]
current_mdl_params={}
for i in mdl_params.keys():
if i.startswith("msm__"):
current_mdl_params[i.split("msm__")[1]] = mdl_params[i]
if "bootstrap__n_samples" in mdl_params.keys():
bootstrap__n_samples = mdl_params["bootstrap__n_samples"]
else:
bootstrap__n_samples = 100
for protein in yaml_file["protein_list"]:
with enter_protein_mdl_dir(yaml_file, protein):
print(protein)
assignments = verboseload("assignments.pkl")
msm_mdl =BootStrapMarkovStateModel(n_samples= bootstrap__n_samples, n_procs=2,
msm_args = current_mdl_params
)
msm_mdl.fit([assignments[i] for i in assignments.keys()], pool=pool)
verbosedump(msm_mdl, "bootstrap_msm_mdl.pkl")
verbosedump(msm_mdl.mle_, "msm_mdl.pkl")
fixed_assignments = {}
for i in assignments.keys():
fixed_assignments[i] = msm_mdl.mle_.transform(
assignments[i], mode='fill')[0]
verbosedump(fixed_assignments, 'fixed_assignments.pkl')
return
def fit_bootstrap(yaml_file,pool=None):
mdl_params = yaml_file["mdl_params"]
current_mdl_params={}
bootstrap_mdl_params={}
for i in mdl_params.keys():
if i.startswith("msm__"):
current_mdl_params[i.split("msm__")[1]] = mdl_params[i]
if i.startswith("bootstrap__"):
bootstrap_mdl_params[i.split("bootstrap__")[1]] = mdl_params[i]
if "n_samples" not in bootstrap_mdl_params.keys():
bootstrap_mdl_params["n_samples"] = 100
for protein in yaml_file["protein_list"]:
with enter_protein_mdl_dir(yaml_file, protein):
print(protein)
assignments = verboseload("assignments.pkl")
msm_mdl =BootStrapMarkovStateModel(n_procs=2,
msm_args = current_mdl_params,
**bootstrap_mdl_params)
msm_mdl.fit([assignments[i] for i in assignments.keys()], pool=pool)
verbosedump(msm_mdl, "bootstrap_msm_mdl.pkl")
verbosedump(msm_mdl.mle_, "msm_mdl.pkl")
fixed_assignments = {}
for i in assignments.keys():
fixed_assignments[i] = msm_mdl.mle_.transform(
assignments[i], mode='fill')[0]
verbosedump(fixed_assignments, 'fixed_assignments.pkl')
return
def test_mle_eq():
seq = [[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 1, 1]]
mle_mdl = MarkovStateModel(lag_time=1)
b_mdl = BootStrapMarkovStateModel(n_samples=10, n_procs=2, msm_args={'lag_time': 1})
mle_mdl.fit(seq)
b_mdl.fit(seq)
#make sure we have good model
eq(mle_mdl.populations_, b_mdl.mle_.populations_)
eq(mle_mdl.timescales_, b_mdl.mle_.timescales_)
def test_resampler():
sequences = [np.random.randint(20, size=100) for _ in range(100)]
mdl = BootStrapMarkovStateModel(n_samples=5, n_procs=2, msm_args={'lag_time': 10})
#probability that
mdl.fit(sequences)
#given a size of 100 input trajectories the probability that
# we re-pick the original set is about (1/100)^100.
# we test that the set of unique traj ids is never equal to
#original 100 sets in all 5 samples
for i in mdl.resample_ind_:
assert len(np.unique(i)) != 100
def test_mle_eq():
seq = [[0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 0, 1, 1]]
mle_mdl = MarkovStateModel(lag_time=1)
b_mdl = BootStrapMarkovStateModel(n_samples=10,
n_procs=2,
msm_args={'lag_time': 1})
mle_mdl.fit(seq)
b_mdl.fit(seq)
#make sure we have good model
eq(mle_mdl.populations_, b_mdl.mle_.populations_)
eq(mle_mdl.timescales_, b_mdl.mle_.timescales_)
def test_score():
seq = [np.random.randint(20, size=100),
np.random.randint(20, size=100),
np.random.randint(20, size=100)]
bmsm = BootStrapMarkovStateModel(n_samples=10, n_procs=2, msm_args={'lag_time':1})
bmsm.fit(seq)
# test that all samples got a training score ...
assert np.array(bmsm.all_training_scores_).shape[0] == 10
# ... and that the training score wasn't NaN
assert sum(np.isnan(bmsm.all_training_scores_)) == 0
# test that a test score was attempted (OK if it's NaN)
assert bmsm.n_samples == np.array(bmsm.all_test_scores_).shape[0]
def test_resampler():
sequences = [np.random.randint(20, size=100) for _ in range(100)]
mdl = BootStrapMarkovStateModel(n_samples=5,
n_procs=2,
msm_args={'lag_time': 10})
#probability that
mdl.fit(sequences)
#given a size of 100 input trajectories the probability that
# we re-pick the original set is about (1/100)^100.
# we test that the set of unique traj ids is never equal to
#original 100 sets in all 5 samples
for i in mdl.resample_ind_:
assert len(np.unique(i)) != 100
def test_score():
seq = [
np.random.randint(20, size=100),
np.random.randint(20, size=100),
np.random.randint(20, size=100)
]
bmsm = BootStrapMarkovStateModel(n_samples=10,
n_procs=2,
msm_args={'lag_time': 1})
bmsm.fit(seq)
# test that all samples got a training score ...
assert np.array(bmsm.all_training_scores_).shape[0] == 10
# ... and that the training score wasn't NaN
assert sum(np.isnan(bmsm.all_training_scores_)) == 0
# test that a test score was attempted (OK if it's NaN)
assert bmsm.n_samples == np.array(bmsm.all_test_scores_).shape[0]
def test_mdl():
mdl = BootStrapMarkovStateModel(n_samples=10,
n_procs=2,
msm_args={'lag_time': 10})
