folder = args.output_folder
file_suffix = args.output_file_suffix
# ----------------------------- SETUP ----------------------------------- #
date = str(datetime.now()).replace(" ", "-").replace(":", '-').replace('.', '-')
log = open("tmp/training_log.txt", "w")
dag = load_go_dag('data/gene_ontology.1_2.obo')
if vectorizer_method not in ['count', 'tf-idf']:
print('Vectorizer Method must select from: count | tf-idf')
sys.exit(1)
if folder:
direc = 'results/{}-{}'.format(folder, date)
su_make_dir(direc)
else:
direc = tempfile.mkdtemp(prefix='{}-{}-'.format(method, date), dir='results/')
selection = []
ontologies = []
if pfam:
selection.append('pfam')
if ipr:
selection.append('ipr')
go_type = None
if induce and go:
go_type = 'induced_go'
elif go and not induce:
go_type = 'go'
labels = sorted(prep.get_labels_from_file('data/labels.tsv'))
train, _ = prep.prep_data_frames(selection=[])
uniprots = list(train['uniprot'].values)
try:
interactome_fp = open(supplied_file, 'r')
except IOError:
print("Could not open supplied file {}.".format(supplied_file))
sys.exit(0)
if label.lower() not in labels:
print("Could not recognise label {}.".format(label))
sys.exit(0)
else:
label = label.lower()
su_make_dir('{}/{}_{}'.format(out_dir, label, threshold))
out_dir = '{}/{}_{}'.format(out_dir, label, threshold)
# Go through interactome and get all reactions with label
header = interactome_fp.readline().strip().split('\t')
protein_a_idx = header.index('p1')
protein_b_idx = header.index('p2')
gene_a_idx = header.index('g1')
gene_b_idx = header.index('g2')
num_annot = header.index('n_annot')
label_idx = header.index(label)
reactions = []
reaction_info = [
('p1', protein_a_idx),
('p2', protein_b_idx),
