def run_AP_clustering_without_local_modules(path, dataset, features, labels):
_, A, _ = load_data(path, dataset, features, labels)
A1 = A.toarray()
for i in range(A1.shape[0]):
A1[i, i] = 0
A1[A1 != 0] = 1
mask = A1
mask[mask != 0] = 0 #set null mask (no penalty will be added to messages)
embeddings = np.genfromtxt("{}{}.embeddings.txt".format(path, dataset),
dtype=np.float32)
print("Clustering by Affinity Propagation ...")
af = AffinityPropagation(damping=0.9,
max_iter=2000,
convergence_iter=20,
copy=False,
preference=None,
affinity='euclidean',
verbose=True,
mask=mask).fit(embeddings)
cluster_centers_indices = af.cluster_centers_indices_
labels = af.labels_
n_clusters_ = len(cluster_centers_indices)
print("Number of obtained clusters: {}".format(n_clusters_))
print("Preparing clustering results ...")
prepare_clusters_AP_without_local_modules(labels, cluster_centers_indices)
print("Clustering results file ready")
print("Preparing enrichment analysis data for PPI network ...")
prepare_enrichment_clusters_without_local_modules(dataset="PPI")
print("Enrichment analysis data ready")
def run_AP_clustering_without_GCN(path, dataset, features, labels):
_, A, _ = load_data(path, dataset, features, labels)
A1 = A.toarray()
for i in range(A1.shape[0]):
A1[i,i] = 0
A1[A1!=0]=1
mask = A1
mask[mask!=0]=1
tmp1 = np.matmul(A1,A1)#A2
tmp = tmp1.copy()
tmp[tmp!=0]=1
mask = np.maximum(mask, tmp)
for i in range(mask.shape[0]):
mask[i,i] = 1
mask[mask==0] = 100
mask[mask==1] = 0
GE_features = features[:, 1:]
print("Clustering by Affinity Propagation ...")
af = AffinityPropagation(damping=0.9, max_iter=2000, convergence_iter=20, copy=False, preference=None, affinity='euclidean', verbose=True, mask=mask).fit(GE_features)
cluster_centers_indices = af.cluster_centers_indices_
labels = af.labels_
n_clusters_ = len(cluster_centers_indices)
print("Number of obtained clusters: {}".format(n_clusters_))
print("Preparing clustering results ...")
prepare_clusters_AP_without_GCN(labels, cluster_centers_indices)
print("Clustering results file ready")
print("Preparing enrichment analysis data for PPI network ...")
prepare_enrichment_clusters_without_GCN(dataset="PPI")
print("Enrichment analysis data ready")