def sample_cv(dataset, model_name, num_folds, fold, use_cosmic, num_signatures,
shuffle_seed, random_seed, max_iterations, epsilon, out_dir):
if fold >= num_folds:
raise ValueError('num_folds is {} but fold is {}'.format(
num_folds, fold))
dataset_name = dataset
dataset, active_signatures = get_data_by_model_name(dataset, model_name)
if use_cosmic:
num_signatures = len(active_signatures)
signatures = get_cosmic_signatures()[active_signatures]
elif num_signatures == 0:
print(
'use_cosmic is False and num_signatures is 0, using number of active cosmic signatures {}'
.format(len(active_signatures)))
num_signatures = len(active_signatures)
signatures = None
else:
signatures = None
use_cosmic_dir = 'refit' if use_cosmic else 'denovo'
out_dir = os.path.join(out_dir, dataset_name, use_cosmic_dir, model_name,
str(num_signatures), str(shuffle_seed),
str(num_folds), str(fold))
try:
os.makedirs(out_dir)
except OSError:
pass
random_seed = int(time.time()) if random_seed == 0 else random_seed
out_file = out_dir + "/" + str(random_seed)
if os.path.isfile(out_file + '.json'):
print(
'Experiment with parameters {} {} {} {} {} {} {} {} already exist'.
format(dataset_name, model_name, num_folds, fold, use_cosmic,
num_signatures, shuffle_seed, random_seed))
return
train_data, test_data = split_train_test_sample_cv(dataset, num_folds,
fold, shuffle_seed)
model, train_ll, test_ll = train_test_stickysig(train_data, test_data,
num_signatures, signatures,
random_seed, epsilon,
max_iterations)
parameters = model.get_params()
parameters['alpha'] = parameters['alpha'].tolist()
parameters['e'] = parameters['e'].tolist()
for sample in parameters['pi']:
parameters['pi'][sample] = parameters['pi'][sample].tolist()
out = {
'log-likelihood-train': train_ll,
'log-likelihood-test': test_ll,
'parameters': parameters
}
save_json(out_file, out)
def leave_one_chromosome_out(dataset, model_name, chromosome, use_cosmic,
num_signatures, random_seed, max_iterations,
epsilon, out_dir):
use_cosmic_dir = 'refit' if use_cosmic else 'denovo'
all_chromosomes = [str(i) for i in range(1, 23)]
all_chromosomes.extend(['X', 'Y'])
chromosome_name = all_chromosomes[chromosome]
dataset_name = dataset
dataset, active_signatures = get_data_by_model_name(dataset, model_name)
if use_cosmic:
num_signatures = len(active_signatures)
signatures = get_cosmic_signatures()[active_signatures]
elif num_signatures == 0:
print(
'use_cosmic is False and num_signatures is 0, using number of active cosmic signatures {}'
.format(len(active_signatures)))
num_signatures = len(active_signatures)
signatures = None
else:
signatures = None
out_dir = os.path.join(out_dir, dataset_name, use_cosmic_dir, model_name,
str(num_signatures), chromosome_name)
try:
os.makedirs(out_dir)
except OSError:
pass
random_seed = int(time.time()) if random_seed == 0 else random_seed
out_file = out_dir + "/" + str(random_seed)
if os.path.isfile(out_file + '.json'):
print('Experiment with parameters {} {} {} {} {} {} already exist'.
format(dataset_name, model_name, chromosome, use_cosmic,
num_signatures, random_seed))
return
train_data, test_data = split_train_test_loco(dataset, chromosome)
model, train_ll, test_ll = train_test_stickysig(train_data, test_data,
num_signatures, signatures,
random_seed, epsilon,
max_iterations)
parameters = model.get_params()
parameters['alpha'] = parameters['alpha'].tolist()
parameters['e'] = parameters['e'].tolist()
for sample in parameters['pi']:
parameters['pi'][sample] = parameters['pi'][sample].tolist()
out = {
'log-likelihood-train': train_ll,
'log-likelihood-test': test_ll,
'parameters': parameters
}
save_json(out_file, out)
def plot_sig_correlations(dataset, num_sigs, beta_loss=2, plot_title=True, save_plot=True):
cosmic_signatures = get_cosmic_signatures()
mix_dir = os.path.join(ROOT_DIR, 'experiments/trained_models/{}/denovo'.format(dataset))
scores_dict = get_best_model(mix_dir, return_params=True)
BIC_scores = scores_dict['BIC_scores']
num_signatures = scores_dict['num_signatures']
model_paths = scores_dict['model_paths']
signatures_dict = {}
# MIX signatures
indices = num_signatures == num_sigs
best_model_path = model_paths[indices][np.argmin(BIC_scores[indices])]
e = np.array(load_json(best_model_path)['parameters']['e'])
signatures_dict['MIX'] = e.copy()
# NMF signatures
data, _ = get_data(dataset)
e = learn_NMF(data, num_sigs, beta_loss=beta_loss)
signatures_dict['NMF'] = e.copy()
# clustered-NMF signatures (if needed)
if dataset == 'MSK-ALL':
data, _ = get_data('clustered-MSK-ALL')
e = learn_NMF(data, num_sigs, beta_loss=beta_loss)
signatures_dict['clustered-NMF'] = e.copy()
plt.rcParams.update({'font.size': 12})
x_axis = np.array([str(i + 1) for i in range(num_sigs)])
plt.axhline(0.80, color='grey', linestyle='--', label='_nolegend_')
legends = []
for i, model in enumerate(signatures_dict.keys()):
legends.append(model)
e = signatures_dict[model]
sigs, corrs = get_signatures_correlations(e, cosmic_signatures)
sigs = sigs[np.argsort(-corrs)]
corrs = corrs[np.argsort(-corrs)]
curr_x_axis = x_axis
color = 'C{}'.format(i)
plt.plot(curr_x_axis, corrs, '.-k', color=color)
for i in range(len(sigs)):
plt.annotate(str(sigs[i] + 1), (i, corrs[i] + 0.002), color=color)
print('{} - {} - {} - {}'.format(model, sigs.tolist(), corrs.tolist(), sum(corrs)))
plt.yticks(np.arange(2, 6) * 0.2)
plt.ylabel('Cosine similarity', fontsize='large')
plt.xlabel('Rank of signature', fontsize='large')
plt.legend(legends, loc='lower left')
if plot_title:
plt.title('{} signatures'.format(num_sigs))
if save_plot:
plt.savefig(os.path.join(ROOT_DIR, 'results', 'signatures_similarity', '{}-signatures.pdf'.format(num_sigs)))
def train_model(dataset, model_name, use_cosmic, num_signatures, random_seed,
max_iterations, epsilon, out_dir):
use_cosmic_dir = 'refit' if use_cosmic else 'denovo'
dataset_name = dataset
dataset, active_signatures = get_data_by_model_name(dataset, model_name)
if use_cosmic:
num_signatures = len(active_signatures)
signatures = get_cosmic_signatures()[active_signatures]
elif num_signatures == 0:
print(
'use_cosmic is False and num_signatures is 0, using number of active cosmic signatures {}'
.format(len(active_signatures)))
num_signatures = len(active_signatures)
signatures = None
else:
signatures = None
out_dir = os.path.join(out_dir, dataset_name, use_cosmic_dir, model_name,
str(num_signatures))
try:
os.makedirs(out_dir)
except OSError:
pass
random_seed = int(time.time()) if random_seed == 0 else random_seed
out_file = out_dir + "/" + str(random_seed)
if os.path.isfile(out_file + '.json'):
print('Experiment with parameters {} {} {} {} {} already exist'.format(
dataset_name, model_name, use_cosmic, num_signatures, random_seed))
return
model, ll = train_stickysig(dataset, num_signatures, signatures,
random_seed, epsilon, max_iterations)
parameters = model.get_params()
parameters['alpha'] = parameters['alpha'].tolist()
parameters['e'] = parameters['e'].tolist()
for sample in parameters['pi']:
parameters['pi'][sample] = parameters['pi'][sample].tolist()
out = {'log-likelihood': ll, 'parameters': parameters}
save_json(out_file, out)
def ROC_HRD():
from src.analyze_results import get_best_model, stack_nnls
import matplotlib.pyplot as plt
from src.utils import get_data, get_cosmic_signatures
import pandas as pd
from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
import seaborn as sns
try:
os.makedirs(os.path.join(ROOT_DIR, 'results', 'HRD'))
except OSError:
pass
# Loading data
test_mutations, signatures = get_data('nature2019-panel')
test_labels, _ = get_data('nature2019-labels')
# fixing to 0, 1. Removing intermediate hrd
test_labels = test_labels[:, 0]
test_data_samples = test_labels != 0
test_labels = test_labels[test_data_samples]
test_labels[test_labels == -1] += 1
# models = ['MIX Sig3', 'SigMA', 'TMB', 'NNLS Sig3', 'WGS Sig3']
models = ['MIX Sig3', 'SigMA', 'NNLS Sig3']
scores = []
for model in models:
if 'MIX Sig3' in model:
# MIX
directory = os.path.join(ROOT_DIR, 'experiments/trained_models/BRCA-panel/refit')
mix = get_best_model(directory, return_model=True)
test_data = mix.weighted_exposures(test_mutations)
if 'normalized' not in model:
test_data *= test_mutations.sum(1, keepdims=True) # un-normalizing exposures
test_data = test_data[:, [2]]
elif model == 'SigMA':
sigma_output = os.path.join(ROOT_DIR, 'data/nature2019/SigMA_output.tsv')
all_df = pd.read_csv(sigma_output, sep='\t')
# In case this is comma separated
if len(all_df.columns) == 1:
all_df = pd.read_csv(sigma_output, sep=',')
test_data = all_df[['Signature_3_mva']].values
# test_data = all_df[['Signature_3_mva']].values
elif model == 'TMB':
test_data = test_mutations.sum(1, keepdims=True)
elif model == 'WGS Sig3':
# NNLS on WGS
a, _ = get_data('nature2019-full')
test_data = stack_nnls(a, get_cosmic_signatures()[signatures])
test_data = test_data[:, [2]]
else:
# NNLS on panel
test_data = stack_nnls(test_mutations, get_cosmic_signatures()[signatures])
test_data = test_data[:, [2]]
test_data = test_data[test_data_samples]
# Test estimator on data
prediction_probabilities = test_data
auc_roc = roc_auc_score(test_labels, prediction_probabilities)
print(model, 'auc: {:.2f}'.format(auc_roc))
scores.append([model, '{:.2f}'.format(auc_roc)])
fpr, tpr, thresholds = roc_curve(test_labels, prediction_probabilities)
sns.lineplot(fpr, tpr, ci=None)
plt.legend(models, loc='lower right')
plt.xlabel('False positive rate')
plt.ylabel('True positive rate')
plt.savefig(os.path.join(ROOT_DIR, 'results', 'HRD', 'ROC_HRD.pdf'))
# plt.show()
np.savetxt(os.path.join(ROOT_DIR, 'results', 'HRD', 'ROC_AUC.tsv'), scores, '%s', '\t')
def ROC_immunotherapy():
from src.analyze_results import get_best_model, stack_nnls, get_signatures_correlations
import matplotlib.pyplot as plt
from src.utils import get_data, get_cosmic_signatures
from sklearn.metrics import roc_curve
from sklearn.metrics import roc_auc_score
import seaborn as sns
try:
os.makedirs(os.path.join(ROOT_DIR, 'results', 'immunotherapy'))
except OSError:
pass
# Loading data
test_mutations, sigs = get_data('msk2018-LUAD')
_, sigs = get_data('MSK-ALL')
test_labels, _ = get_data('msk2018-LUAD-labels')
models = ['MIX refit Sig4', 'MIX denovo M5', 'TMB', 'NNLS Sig4']
scores = []
for model in models:
if 'MIX refit Sig4' in model:
# MIX
directory = os.path.join(ROOT_DIR, 'experiments/trained_models/MSK-ALL/refit')
mix = get_best_model(directory, return_model=True)
test_data = mix.weighted_exposures(test_mutations)
if 'normalized' not in model:
test_data *= test_mutations.sum(1, keepdims=True) # un-normalizing exposures
test_data = test_data[:, [3]]
elif 'MIX denovo M5' in model:
# MIX
directory = os.path.join(ROOT_DIR, 'experiments/trained_models/MSK-ALL/denovo')
mix = get_best_model(directory, return_model=True)
test_data = mix.weighted_exposures(test_mutations)
if 'normalized' not in model:
test_data *= test_mutations.sum(1, keepdims=True) # un-normalizing exposures
sig4 = get_cosmic_signatures()[sigs][[3]]
_, corrs = get_signatures_correlations(mix.e, sig4)
best_sig = np.argmax(corrs)
test_data = test_data[:, [best_sig]]
elif model == 'TMB':
test_data = test_mutations.sum(1, keepdims=True)
else:
# NNLS
test_data = stack_nnls(test_mutations, get_cosmic_signatures()[sigs])
test_data = test_data[:, [3]]
# Test estimator on data
prediction_probabilities = test_data
auc_roc = roc_auc_score(test_labels, prediction_probabilities)
print(model, 'auc: {:.2f}'.format(auc_roc))
scores.append([model, '{:.2f}'.format(auc_roc)])
fpr, tpr, thresholds = roc_curve(test_labels, prediction_probabilities)
sns.lineplot(fpr, tpr, ci=None)
plt.legend(models, loc='lower right')
plt.xlabel('False positive rate')
plt.ylabel('True positive rate')
plt.savefig(os.path.join(ROOT_DIR, 'results', 'immunotherapy', 'ROC_immunotherapy.pdf'))
# plt.show()
np.savetxt(os.path.join(ROOT_DIR, 'results', 'immunotherapy', 'ROC_AUC.tsv'), scores, '%s', '\t')
def compare_panel_clusters():
np.random.seed(1359)
# full_dataset, panel_dataset = 'BRCA-panel-full', 'BRCA-panel'
full_dataset, panel_dataset = 'nature2019-full', 'nature2019-panel'
full_data, active_signatures = get_data(full_dataset)
panel_data, _ = get_data(panel_dataset)
signatures = get_cosmic_signatures()[active_signatures]
num_samples = len(full_data)
full_data_exposures = stack_nnls(full_data, signatures)
full_data_exposures_dists = cosine_similarity(full_data_exposures)
corrs = []
models = []
relations = []
for model in ['MIX', 'SigMA']:
if model == 'MIX':
d = os.path.join(ROOT_DIR, 'experiments/trained_models/{}/refit'.format('BRCA-panel'))
mix = get_model(load_json(get_best_model(d))['parameters'])
clusters = np.argmax(mix.soft_cluster(full_data), 1)
elif model == 'SigMA':
# d = os.path.join(ROOT_DIR, 'data/ICGC-BRCA/out-sigma-brca-panel.tsv')
d = os.path.join(ROOT_DIR, 'data/nature2019/SigMA_output.tsv')
all_df = pd.read_csv(d, sep='\t')
# In case this is comma separated
if len(all_df.columns) == 1:
all_df = pd.read_csv(d, sep=',')
clusters = all_df['categ'].values
unique_clusters = np.unique(clusters)
cluster_to_num = {}
for i, c in enumerate(unique_clusters):
cluster_to_num[c] = i
clusters = np.array([cluster_to_num[c] for c in clusters])
else:
raise ValueError('error')
dists_in_clusters = []
dists_out_clusters = []
for i in range(num_samples):
for j in range(i + 1, num_samples):
if clusters[i] == clusters[j]:
dists_in_clusters.append(full_data_exposures_dists[i, j])
else:
dists_out_clusters.append(full_data_exposures_dists[i, j])
dists_in_clusters = np.array(dists_in_clusters)
dists_out_clusters = np.array(dists_out_clusters)
dists_in_clusters = np.random.choice(dists_in_clusters, 200, replace=False)
dists_out_clusters = np.random.choice(dists_out_clusters, 200, replace=False)
corrs.extend(dists_in_clusters)
corrs.extend(dists_out_clusters)
models.extend([model] * len(dists_out_clusters) * 2)
relations.extend(['Intra-cluster pairs'] * len(dists_out_clusters))
relations.extend(['Inter-cluster pairs'] * len(dists_out_clusters))
print(model, len(np.unique(clusters)))
print(ranksums(dists_in_clusters, dists_out_clusters), np.mean(dists_in_clusters), np.mean(dists_out_clusters))
df = {'Cosine similarity': corrs, 'model': models, 'relation': relations}
df = pd.DataFrame(df)
sns.violinplot(x='relation', y='Cosine similarity', hue='model', data=df, split=True, inner='stick')
plt.xlabel('')
plt.savefig(os.path.join(ROOT_DIR, 'results', 'clusters_quality', 'clusters_quality.pdf'))
def RE(dataset, models=None, computation='mutations'):
# Handle input
if models is None:
models = ['MIX-conditional-clustering', 'MIX-soft-clustering', 'NNLS']
if dataset == 'BRCA':
full_dataset_name = 'ICGC-BRCA'
elif dataset == 'OV':
full_dataset_name = 'TCGA-OV'
else:
raise ValueError('{} is no a valid dataset'.format(dataset))
if computation == 'mutations':
RE_func = compute_RE_per_sample
elif computation == 'exposures':
RE_func = compute_exposures_RE_per_sample
else:
raise ValueError('{} is no a valid computation inpute'.format(computation))
# Prepare data
full_data, active_signatures = get_data(full_dataset_name)
normalized_full_data = full_data / full_data.sum(1, keepdims=1)
full_panel_data, _ = get_data('{}-panel-full'.format(dataset))
normalized_full_panel_data = full_panel_data / full_panel_data.sum(1, keepdims=1)
signatures = get_cosmic_signatures()[active_signatures]
trained_model_dir = os.path.join(ROOT_DIR, 'experiments/trained_models')
all_experiments = os.listdir(trained_model_dir)
results = {s.split('-par')[0]: {} for s in all_experiments if dataset in s and 'part' in s}
for ds_dataset_name in results:
results[ds_dataset_name] = {model: [] for model in models}
# Prepare data
ds_dataset_part1_name = ds_dataset_name + '-part' + str(1)
ds_dataset_part2_name = ds_dataset_name + '-part' + str(2)
ds_data_part1, _ = get_data(ds_dataset_part1_name)
ds_data_part2, _ = get_data(ds_dataset_part2_name)
ds_data = np.row_stack((ds_data_part1, ds_data_part2))
if 'panel' in ds_dataset_name:
curr_normalized_full_data = normalized_full_panel_data
else:
curr_normalized_full_data = normalized_full_data
# Find the best model
mix_part1 = get_best_model(os.path.join(trained_model_dir, ds_dataset_part1_name, 'refit'), return_model=True)
mix_part2 = get_best_model(os.path.join(trained_model_dir, ds_dataset_part2_name, 'refit'), return_model=True)
# MIX RE with cluster's pi using conditional probability
if 'MIX-conditional-clustering' in models:
clusters = np.argmax(mix_part1.soft_cluster(ds_data_part2), axis=1)
exposures_part2 = mix_part1.pi[clusters]
clusters = np.argmax(mix_part2.soft_cluster(ds_data_part1), axis=1)
exposures_part1 = mix_part2.pi[clusters]
exposures = np.row_stack((exposures_part1, exposures_part2))
results[ds_dataset_name]['MIX-conditional-clustering'] = \
RE_func(curr_normalized_full_data, exposures, signatures)
# MIX RE with weighted cluster pi
if 'MIX-soft-clustering' in models:
exposures_part2 = mix_part1.weighted_exposures(ds_data_part2)
exposures_part1 = mix_part2.weighted_exposures(ds_data_part1)
exposures = np.row_stack((exposures_part1, exposures_part2))
results[ds_dataset_name]['MIX-soft-clustering'] = \
RE_func(curr_normalized_full_data, exposures, signatures)
# NNLS RE
if 'NNLS' in models:
exposures = []
for m in ds_data:
exposures.append(nnls(signatures.T, m)[0])
exposures = np.array(exposures)
results[ds_dataset_name]['NNLS'] = RE_func(curr_normalized_full_data, exposures, signatures)
if 'SigMA' in models:
if 'panel' in ds_dataset_name:
path = 'data/sigma_exposure/out-sigma-{}-panel-full.tsv'.format(dataset.lower())
else:
ds = ds_dataset_name.split('ds')[-1]
path = 'data/sigma_exposure/out-sigma-{}-downsize{}.tsv'.format(dataset.lower(), ds.zfill(3))
# exposures = sigma_output_to_exposures(path)[:, active_signatures]
# print('Active_signatures: {}'.format(np.where(sigma_output_to_exposures(path).sum(0) > 0)[0] + 1))
exposures = sigma_output_to_exposures(path)
exposures /= exposures.sum(1, keepdims=True)
results[ds_dataset_name]['SigMA'] = RE_func(curr_normalized_full_data, exposures, get_cosmic_signatures())
# Shallow analysis (no p-values)
summed_RE_results = {}
for s in results:
summed_RE_results[s] = {}
for m in results[s]:
summed_RE_results[s][m] = np.sum(results[s][m]) / len(results[s][m])
df = pd.DataFrame(summed_RE_results).T
return df
def plot_cluster_AMI(range_clusters, computation='AMI'):
if computation == 'AMI':
score_func = AMI_score
elif computation == 'MI':
score_func = MI_score
elif computation == 'jaccard':
score_func = Jaccard_score
else:
raise ValueError('{} is not a valid computation'.format(computation))
rich_sample_threshold = 10
data, active_signatures = get_data('MSK-ALL')
signatures = get_cosmic_signatures()[active_signatures]
num_data_points = data.sum()
nnls_exposures = np.zeros((len(data), len(signatures)))
for i in range(len(data)):
nnls_exposures[i] = nnls(signatures.T, data[i])[0]
num_mutations_per_sample = data.sum(1)
rich_samples = num_mutations_per_sample >= rich_sample_threshold
all_df = pd.read_csv(os.path.join(ROOT_DIR, 'data/MSK-processed/oncotype_counts.txt'), sep='\t')
all_df['Counts'] = all_df['Counts'].astype(int)
all_df = all_df[all_df['Counts'] > 100]
cancer_types = np.array(all_df['Oncotree'])
sample_cancer_assignments = []
sample_cancer_id_assignments = []
for i, oc in enumerate(cancer_types):
# dat_f = "data/processed/%s_counts.npy" % oc
dat_f = os.path.join(ROOT_DIR, 'data/MSK-processed/{}_counts.npy'.format(oc))
tmp_data = np.array(np.load(dat_f, allow_pickle=True), dtype=np.float64)
sample_cancer_assignments.extend([oc] * len(tmp_data))
sample_cancer_id_assignments.extend([i] * len(tmp_data))
sample_cancer_assignments = np.array(sample_cancer_assignments)
sample_cancer_id_assignments = np.array(sample_cancer_id_assignments)
shuffled_indices = np.arange(len(sample_cancer_assignments))
# Finding best_models
d = os.path.join(ROOT_DIR, 'experiments/trained_models/MSK-ALL/denovo')
BIC_summary = get_best_model(d, return_params=True)
BIC_scores, BIC_clusters, BIC_paths = BIC_summary['BIC_scores'], BIC_summary['num_clusters'], BIC_summary['model_paths']
MIX_scores = np.zeros((2, len(range_clusters)))
MIX_soft_scores = np.zeros((2, len(range_clusters)))
MIX_refit_scores = np.zeros((2, len(range_clusters)))
MIX_soft_refit_scores = np.zeros((2, len(range_clusters)))
KMeans_scores = np.zeros((2, len(range_clusters)))
NNLS_KMeans_scores = np.zeros((2, len(range_clusters)))
for idx, num_clusters in enumerate(range_clusters):
best_model_path = BIC_paths[BIC_clusters == num_clusters][np.argmin(BIC_scores[BIC_clusters == num_clusters])]
model = get_model(load_json(best_model_path)['parameters'])
MIX_soft_clustering = model.soft_cluster(data)
sample_cluster_assignment_MIX = np.argmax(MIX_soft_clustering, 1)
MIX_scores[0, idx] = score_func(sample_cancer_id_assignments, sample_cluster_assignment_MIX)
MIX_scores[1, idx] = score_func(sample_cancer_id_assignments[rich_samples],
sample_cluster_assignment_MIX[rich_samples])
if computation == 'MI':
MIX_soft_scores[0, idx] = MI_score_soft_clustering(sample_cancer_id_assignments, MIX_soft_clustering)
MIX_soft_scores[1, idx] = MI_score_soft_clustering(sample_cancer_id_assignments[rich_samples],
MIX_soft_clustering[rich_samples])
# MIX refit
d = os.path.join(ROOT_DIR, 'experiments/trained_models/MSK-ALL/refit/mix_{}clusters_017signatures'.format(str(num_clusters).zfill(3)))
model = get_model(load_json(get_best_run(d))['parameters'])
MIX_refit_soft_clustering = model.soft_cluster(data)
sample_cluster_assignment_MIX_refit = np.argmax(MIX_refit_soft_clustering, 1)
MIX_refit_scores[0, idx] = score_func(sample_cancer_id_assignments, sample_cluster_assignment_MIX_refit)
MIX_refit_scores[1, idx] = score_func(sample_cancer_id_assignments[rich_samples],
sample_cluster_assignment_MIX_refit[rich_samples])
if computation == 'MI':
MIX_soft_refit_scores[0, idx] = MI_score_soft_clustering(sample_cancer_id_assignments, MIX_refit_soft_clustering)
MIX_soft_refit_scores[1, idx] = MI_score_soft_clustering(sample_cancer_id_assignments[rich_samples],
MIX_refit_soft_clustering[rich_samples])
# KMeans clustering
cluster_model = KMeans(num_clusters, n_init=100, random_state=140296)
np.random.shuffle(shuffled_indices)
shuffled_data = data[shuffled_indices]
cluster_model.fit(shuffled_data)
kmeans_clusters = cluster_model.predict(data)
KMeans_scores[0, idx] = score_func(sample_cancer_id_assignments, kmeans_clusters)
KMeans_scores[1, idx] = score_func(sample_cancer_id_assignments[rich_samples],
kmeans_clusters[rich_samples])
# NNLS + KMeans clustering
cluster_model = KMeans(num_clusters, n_init=100, random_state=140296)
np.random.shuffle(shuffled_indices)
shuffled_data = nnls_exposures[shuffled_indices]
cluster_model.fit(shuffled_data)
nnls_kmeans_clusters = cluster_model.predict(nnls_exposures)
NNLS_KMeans_scores[0, idx] = score_func(sample_cancer_id_assignments, nnls_kmeans_clusters)
NNLS_KMeans_scores[1, idx] = score_func(sample_cancer_id_assignments[rich_samples],
nnls_kmeans_clusters[rich_samples])
print('finished {}'.format(num_clusters))
plt.plot(range_clusters, MIX_scores[0], label='MIX-denovo')
if computation == 'MI':
plt.plot(range_clusters, MIX_soft_scores[0], label='MIX-denovo-soft')
plt.plot(range_clusters, MIX_refit_scores[0], label='MIX-refit')
if computation == 'MI':
plt.plot(range_clusters, MIX_soft_refit_scores[0], label='MIX-refit-soft')
plt.plot(range_clusters, KMeans_scores[0], label='KMeans')
plt.plot(range_clusters, NNLS_KMeans_scores[0], label='NNLS+KMeans')
plt.title('All samples AMI score')
plt.xlabel('clusters')
plt.ylabel(computation)
plt.legend(loc='lower right')
plt.xticks(np.arange(min(range_clusters), max(range_clusters) + 1, 2))
plt.savefig(os.path.join(ROOT_DIR, 'results', 'AMI', 'cluster_score_all.pdf'))
# plt.show()
plt.plot(range_clusters, MIX_scores[1], label='MIX-denovo')
if computation == 'MI':
plt.plot(range_clusters, MIX_soft_scores[1], label='MIX-denovo-soft')
plt.plot(range_clusters, MIX_refit_scores[1], label='MIX-refit')
if computation == 'MI':
plt.plot(range_clusters, MIX_soft_refit_scores[1], label='MIX-refit-soft')
plt.plot(range_clusters, KMeans_scores[1], label='KMeans')
plt.plot(range_clusters, NNLS_KMeans_scores[1], label='NNLS+KMeans')
plt.title('Filtered AMI score')
plt.xlabel('clusters')
plt.ylabel(computation)
plt.legend(loc='lower right')
plt.xticks(np.arange(min(range_clusters), max(range_clusters) + 1, 2))
plt.savefig(os.path.join(ROOT_DIR, 'results', 'AMI', 'cluster_score_filtered.pdf'))
# plt.show()
return
