def main(argv):
try:
opts, args = getopt.getopt(argv, "m:d:f:c:s:o:n:p", ["method=", "dataset=", "data-dir=", "cvs=", "specify-arg=", "method-options=", "predict-num=", "output-dir=", ])
except getopt.GetoptError:
sys.exit()
data_dir = os.path.join(os.path.pardir, 'data')
output_dir = os.path.join(os.path.pardir, 'output')
cvs, sp_arg, model_settings, predict_num = 1, 1, [], 0
seeds = [7771, 8367, 22, 1812, 4659]
# seeds = np.random.choice(10000, 5, replace=False)
for opt, arg in opts:
if opt == "--method":
method = arg
if opt == "--dataset":
dataset = arg
if opt == "--data-dir":
data_dir = arg
if opt == "--output-dir":
output_dir = arg
if opt == "--cvs":
cvs = int(arg)
if opt == "--specify-arg":
sp_arg = int(arg)
if opt == "--method-options":
model_settings = [s.split('=') for s in str(arg).split()]
if opt == "--predict-num":
predict_num = int(arg)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
# default parameters for each methods
if method == 'nrlmf':
args = {'c': 5, 'K1': 5, 'K2': 5, 'r': 50, 'lambda_d': 0.125, 'lambda_t': 0.125, 'alpha': 0.25, 'beta': 0.125, 'theta': 0.5, 'max_iter': 100}
# args = {'c': 5, 'K1': 2, 'K2': 2, 'r': 3, 'lambda_d': 0.125, 'lambda_t': 0.125, 'alpha': 0.25, 'beta': 0.125, 'theta': 0.5, 'max_iter': 100}
if method == 'netlaprls':
args = {'gamma_d': 10, 'gamma_t': 10, 'beta_d': 1e-5, 'beta_t': 1e-5}
if method == 'blmnii':
args = {'alpha': 0.7, 'gamma': 1.0, 'sigma': 1.0, 'avg': False}
if method == 'wnngip':
args = {'T': 0.8, 'sigma': 1.0, 'alpha': 0.8}
if method == 'kbmf':
args = {'R': 50}
if method == 'cmf':
args = {'K': 50, 'lambda_l': 0.5, 'lambda_d': 0.125, 'lambda_t': 0.125, 'max_iter': 30}
for key, val in model_settings:
args[key] = val
#ZINC test lines
intMat, testMat, drugMat, targetMat = load_data_from_file_csv(dataset, os.path.join(data_dir))
drug_names, target_names = get_drug_target_names_zinc()
#ZINC test lines
# drug_names, target_names = get_drugs_targets_names(dataset, os.path.join(data_dir))
#Demo lines
# intMat, testMat, drugMat, targetMat = load_data_from_file_demo(dataset, os.path.join(data_dir))
# drug_names, target_names = get_drug_target_names_demo()
#demo lines
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv = intMat.T, targetMat, drugMat, 0
cv_data = cross_validation(X, seeds, cv)
if sp_arg == 0 and predict_num == 0:
if method == 'nrlmf':
cv_eval.nrlmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'kbmf':
cv_eval.kbmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if sp_arg == 1 or predict_num > 0:
tic = time.clock()
if method == 'nrlmf':
model = NRLMF(cfix=args['c'], K1=args['K1'], K2=args['K2'], num_factors=args['r'], lambda_d=args['lambda_d'], lambda_t=args['lambda_t'], alpha=args['alpha'], beta=args['beta'], theta=args['theta'], max_iter=args['max_iter'])
if method == 'netlaprls':
model = NetLapRLS(gamma_d=args['gamma_d'], gamma_t=args['gamma_t'], beta_d=args['beta_t'], beta_t=args['beta_t'])
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'], gamma=args['gamma'], sigma=args['sigma'], avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=args['T'], sigma=args['sigma'], alpha=args['alpha'])
if method == 'kbmf':
model = KBMF(num_factors=args['R'])
if method == 'cmf':
model = CMF(K=args['K'], lambda_l=args['lambda_l'], lambda_d=args['lambda_d'], lambda_t=args['lambda_t'], max_iter=args['max_iter'])
cmd = str(model)
if predict_num == 0:
print "Dataset:"+dataset+" CVS:"+str(cvs)+"\n"+cmd
aupr_vec, auc_vec = train(model, cv_data, X, D, T)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
print "auc:%.6f, aupr: %.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f" % (auc_avg, aupr_avg, auc_conf, aupr_conf, time.clock()-tic)
write_metric_vector_to_file(auc_vec, os.path.join(output_dir, method+"_auc_cvs"+str(cvs)+"_"+dataset+".txt"))
write_metric_vector_to_file(aupr_vec, os.path.join(output_dir, method+"_aupr_cvs"+str(cvs)+"_"+dataset+".txt"))
elif predict_num > 0:
print "Dataset:"+dataset+"\n"+cmd
seed = 7771 if method == 'cmf' else 22
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
# x, y = np.where(intMat == 0)
x, y = np.where(intMat >= 0) #to pick all pairs including train pairs
scores = model.predict_scores_NaNtest(zip(x, y), 5)
def thear(method, dataset, data_dir, output_dir, cvs, sp_arg, model_settings,
predict_num, seeds, seedsOptPar, args):
intMat, drugMat, targetMat = load_data_from_file(
dataset, os.path.join(data_dir, 'datasets'))
drug_names, target_names = get_drugs_targets_names(
dataset, os.path.join(data_dir, 'datasets'))
invert = 0
if (method == 'inv_brdti'):
invert = 1
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv, invert = intMat.T, targetMat, drugMat, 0, 1
if cvs == 4:
X, D, T, cv = intMat, drugMat, targetMat, 2
cv_data = cross_validation(X, seeds, cv, invert, num=10)
if invert:
X, D, T = intMat, drugMat, targetMat
#cv_data_optimize_params = cross_validation(X, seedsOptPar, cv, invert, num=5)
if sp_arg == 0 and predict_num == 0:
if (method == "vbmklmf"):
cv_eval.vbmklmf_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if (method == "ensambledti"):
cv_eval.vbmklmf_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'netcbp':
cv_eval.netcbp_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
# if method == 'ndaf':
# cv_eval.ndaf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'grmf':
cv_eval.grmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'pudt':
cv_eval.pudt_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'daspfind':
cv_eval.daspfind_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'dnilmf':
cv_eval.dnilmf_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'dthybrid':
cv_eval.dthybrid_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'kronrlsmkl':
cv_eval.kronrismkl_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if (method == 'brdti'):
cv_eval.brdti_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if (method == 'ddr'):
cv_eval.ddr_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if (method == 'brdti'):
cv_eval.brdti_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if (method == 'inv_brdti'):
cv_eval.brdti_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'nrlmf':
cv_eval.nrlmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'kbmf':
cv_eval.kbmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if sp_arg == 1 or predict_num > 0:
tic = time.clock()
if (method == "netcbp"):
model = NetCBP()
# if (method=="ndaf"):
# model = NDAF()
if (method == "grmf"):
model = GRMF(cv=cvs)
if (method == "pudt"):
model = PUDT(dataset=dataset)
if (method == "vbmklmf"):
model = VBMKLMF(dataset=dataset, cvs=cvs)
if (method == 'dnilmf'):
model = DNILMF(dataset=dataset)
if (method == 'kronrlsmkl'):
model = KronRLsMKL(dataset=dataset)
if (method == 'dthybrid'):
model = DTHYBRID(dataset=dataset)
if (method == 'daspfind'):
model = DASPFIND(alpha=args['alpha'])
if (method == 'brdti') | (method == 'inv_brdti'):
#model = BRDTI(D=args['D'],learning_rate= args['learning_rate'],max_iters=args['max_iters'],simple_predict=args['simple_predict'],bias_regularization=args['bias_regularization'],global_regularization=args['global_regularization'],cbSim=args['cbSim'],cb_alignment_regularization_user=args['cb_alignment_regularization_user'],cb_alignment_regularization_item=args['cb_alignment_regularization_item'])
model = BRDTI(args)
if method == 'nrlmf':
model = NRLMF(cfix=args['c'],
K1=args['K1'],
K2=args['K2'],
num_factors=args['r'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
alpha=args['alpha'],
beta=args['beta'],
theta=args['theta'],
max_iter=args['max_iter'])
if method == 'ddr':
model = DDR(dataset=dataset, cv=cvs)
if method == 'netlaprls':
model = NetLapRLS(gamma_d=args['gamma_d'],
gamma_t=args['gamma_t'],
beta_d=args['beta_t'],
beta_t=args['beta_t'])
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'],
gamma=args['gamma'],
sigma=args['sigma'],
avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=args['T'],
sigma=args['sigma'],
alpha=args['alpha'])
if method == 'kbmf':
model = KBMF(num_factors=args['R'])
if method == 'cmf':
model = CMF(K=args['K'],
lambda_l=args['lambda_l'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
max_iter=args['max_iter'])
if (method == 'ensambledti'):
model = EnsambleDTI(args=args, dataset=dataset)
cmd = str(model)
if predict_num == 0:
print("Dataset:" + dataset + " CVS:" + str(cvs) + "\n" + cmd)
aupr_vec, auc_vec = train(model, cv_data, X, D, T, cvs, dataset)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
print(
"auc:%.6f, aupr: %.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f"
% (auc_avg, aupr_avg, auc_conf, aupr_conf, time.clock() - tic))
write_metric_vector_to_file(
auc_vec,
os.path.join(
output_dir,
method + "_auc_cvs" + str(cvs) + "_" + dataset + ".txt"))
write_metric_vector_to_file(
aupr_vec,
os.path.join(
output_dir,
method + "_aupr_cvs" + str(cvs) + "_" + dataset + ".txt"))
elif predict_num > 0:
print("Dataset:" + dataset + "\n" + cmd)
seed = 7771 if method == 'cmf' else 22
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
x, y = np.where(intMat == 0)
scores = model.predict_scores(zip(x, y), 5)
ii = np.argsort(scores)[::-1]
predict_pairs = [(drug_names[x[i]], target_names[y[i]], scores[i])
for i in ii[:predict_num]]
new_dti_file = os.path.join(
output_dir, "_".join([method, dataset, "new_dti.txt"]))
novel_prediction_analysis(predict_pairs, new_dti_file,
os.path.join(data_dir, 'biodb'))
def learn_hyperparameters(self, intMat, drugMat, targetMat, seed=500):
cv_data_optimize_params = cross_validation(intMat, [seed], 1, 0, num=5)
params = cv_eval.blmnii_cv_eval("blmnii", "", "",
cv_data_optimize_params, intMat,
drugMat, targetMat, 1, "")
self.alpha = params["alpha"]
def main(argv):
try:
opts, args = getopt.getopt(argv, m:d:f:c:s:o:n:p, ["method=", "dataset=", "data-dir=", "cvs=", "specify-arg=", "method-options=", "predict-num=", "output-dir=", ])
except getopt.GetoptError:
sys.exit()
data_dir = 'data'
output_dir = 'output'
cvs, sp_arg, model_settings, predict_num = 1, 1, [], 0
seeds = [7771, 8367, 22, 1812, 4659]
seedsOptPar = [156]
# seeds = np.random.choice(10000, 5, replace=False)
for opt, arg in opts:
if opt == "--method":
method = arg
if opt == "--dataset":
dataset = arg
if opt == "--data-dir":
data_dir = arg
if opt == "--output-dir":
output_dir = arg
if opt == "--cvs":
cvs = int(arg)
if opt == "--specify-arg":
sp_arg = int(arg)
if opt == "--method-options":
model_settings = [s.split('=') for s in str(arg).split()]
if opt == "--predict-num":
predict_num = int(arg)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
if not os.path.isdir(os.path.join(output_dir,"optPar")):
os.makedirs(os.path.join(output_dir,"optPar"))
# default parameters for each methods
if (method == 'brdti') | (method == 'inv_brdti') :
args = {
'D':100,
'learning_rate':0.1,
'max_iters' : 100,
'simple_predict' :False,
'bias_regularization':1,
'global_regularization':10**(-2),
"cbSim": "knn",
'cb_alignment_regularization_user' :1,
'cb_alignment_regularization_item' :1}
if method == 'netlaprls':
args = {'gamma_d': 10, 'gamma_t': 10, 'beta_d': 1e-5, 'beta_t': 1e-5}
if method == 'blmnii':
args = {'alpha': 0.7, 'gamma': 1.0, 'sigma': 1.0, 'avg': False}
if method == 'wnngip':
args = {'T': 0.8, 'sigma': 1.0, 'alpha': 0.8}
if method == 'cmf':
args = {'K': 100, 'lambda_l': 0.5, 'lambda_d': 0.125, 'lambda_t': 0.125, 'max_iter': 100}
#print(model_settings)
for key, val in model_settings:
args[key] = float(eval(val))
intMat, drugMat, targetMat = load_data_from_file(dataset, os.path.join(data_dir, 'datasets'))
drug_names, target_names = get_drugs_targets_names(dataset, os.path.join(data_dir, 'datasets'))
invert = 0
if (method == 'inv_brdti') :
invert = 1
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv = intMat.T, targetMat, drugMat, 0
cv_data = cross_validation(X, seeds, cv, invert)
cv_data_optimize_params = cross_validation(X, seedsOptPar, cv, invert, num=5)
if sp_arg == 0 and predict_num == 0:
if (method == 'brdti'):
cv_eval.brdti_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X, D, T, cvs, args)
if (method == 'inv_brdti'):
cv_eval.brdti_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X.T, T, D, cvs, args)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X, D, T, cvs, args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X, D, T, cvs, args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset,output_dir, cv_data_optimize_params, X, D, T, cvs, args)
if sp_arg == 1 or predict_num > 0:
tic = time.clock()
if (method == 'brdti')|(method == 'inv_brdti'):
model = BRDTI(args)
if method == 'netlaprls':
model = NetLapRLS(gamma_d=args['gamma_d'], gamma_t=args['gamma_t'], beta_d=args['beta_t'], beta_t=args['beta_t'])
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'], gamma=args['gamma'], sigma=args['sigma'], avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=args['T'], sigma=args['sigma'], alpha=args['alpha'])
if method == 'cmf':
model = CMF(K=args['K'], lambda_l=args['lambda_l'], lambda_d=args['lambda_d'], lambda_t=args['lambda_t'], max_iter=args['max_iter'])
cmd = str(model)
#predict hidden part of the current datasets
if predict_num == 0:
print "Dataset:"+dataset+" CVS:"+str(cvs)+"\n"+cmd
if (method == 'inv_brdti') :
aupr_vec, auc_vec, ndcg_inv_vec, ndcg_vec, results = train(model, cv_data, X.T, T, D)
else:
aupr_vec, auc_vec, ndcg_vec, ndcg_inv_vec, results = train(model, cv_data, X, D, T)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
ndcg_avg, ndcg_conf = mean_confidence_interval(ndcg_vec)
ndcg_inv_avg, ndcg_inv_conf = mean_confidence_interval(ndcg_inv_vec)
resfile = os.path.join('output','rawResults', method+"_res_"+str(cvs)+"_"+dataset+".csv")
outd = open(resfile, "w")
outd.write(('drug;target;true;predict\n'))
for r in results:
outd.write('%s;%s;%s;%s\n' % (r[0],r[1],r[2],r[3]) )
print "auc:%.6f, aupr: %.6f, ndcg: %.6f, ndcg_inv: %.6f, auc_conf:%.6f, aupr_conf:%.6f, ndcg_conf:%.6f, ndcg_inv_conf:%.6f, Time:%.6f" % (auc_avg, aupr_avg, ndcg_avg, ndcg_inv_avg, auc_conf, aupr_conf, ndcg_conf, ndcg_inv_conf, time.clock()-tic)
write_metric_vector_to_file(auc_vec, os.path.join(output_dir, method+"_auc_cvs"+str(cvs)+"_"+dataset+".txt"))
write_metric_vector_to_file(aupr_vec, os.path.join(output_dir, method+"_aupr_cvs"+str(cvs)+"_"+dataset+".txt"))
write_metric_vector_to_file(ndcg_vec, os.path.join(output_dir, method+"_ndcg_cvs"+str(cvs)+"_"+dataset+".txt"))
write_metric_vector_to_file(ndcg_inv_vec, os.path.join(output_dir, method+"_ndcg_inv_cvs"+str(cvs)+"_"+dataset+".txt"))
#predict novel DTIs
elif predict_num > 0:
print "Dataset:"+dataset+"\n"+cmd
seed = 376
if invert: #predicting drugs for targets
model.fix_model(intMat.T, intMat.T, targetMat, drugMat, seed)
npa = newDTIPrediction()
x, y = np.where(intMat == 0)
scores = model.predict_scores(zip(y, x), 1)
sz = np.array(zip(x,y,scores))
else: #predicting targets for drugs
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
npa = newDTIPrediction()
x, y = np.where(intMat == 0)
scores = model.predict_scores(zip(x, y), 1)
sz = np.array(zip(x,y,scores))
ndcg_d, ndcg_t, recall_d, recall_t = npa.verify_novel_interactions(method, dataset, sz, predict_num, drug_names, target_names)
st_file= os.path.join('output/newDTI', "_".join([dataset,str(predict_num), "stats.csv"]))
out = open(st_file, "a")
out.write(('%s;%f;%f;%f;%f\n' % (method,ndcg_d, ndcg_t, recall_d, recall_t)))
def main(argv):
try:
opts, args = getopt.getopt(argv, "m:d:f:c:s:o:n:p", [
"method=",
"dataset=",
"data-dir=",
"cvs=",
"specify-arg=",
"method-options=",
"predict-num=",
"output-dir=",
])
except getopt.GetoptError:
sys.exit()
data_dir = os.path.join(os.path.pardir, 'data')
output_dir = os.path.join(os.path.pardir, 'output')
cvs, sp_arg, model_settings, predict_num = 1, 1, [], 0
seeds = [7771, 8367, 22, 1812, 4659]
# seeds = np.random.choice(10000, 5, replace=False)
for opt, arg in opts:
if opt == "--method":
method = arg
if opt == "--dataset":
dataset = arg
if opt == "--data-dir":
data_dir = arg
if opt == "--output-dir":
output_dir = arg
if opt == "--cvs":
cvs = int(arg)
if opt == "--specify-arg":
sp_arg = int(arg)
if opt == "--method-options":
model_settings = [s.split('=') for s in str(arg).split()]
if opt == "--predict-num":
predict_num = int(arg)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
# default parameters for each methods
if method == 'nrlmf':
args = {
'c': 5,
'K1': 5,
'K2': 5,
'r': 50,
'lambda_d': 0.125,
'lambda_t': 0.125,
'alpha': 0.25,
'beta': 0.125,
'theta': 0.5,
'max_iter': 100
}
if method == 'netlaprls':
args = {'gamma_d': 10, 'gamma_t': 10, 'beta_d': 1e-5, 'beta_t': 1e-5}
if method == 'blmnii':
args = {'alpha': 0.7, 'gamma': 1.0, 'sigma': 1.0, 'avg': False}
if method == 'wnngip':
args = {'T': 0.8, 'sigma': 1.0, 'alpha': 0.8}
if method == 'kbmf':
args = {'R': 50}
if method == 'cmf':
args = {
'K': 50,
'lambda_l': 0.5,
'lambda_d': 0.125,
'lambda_t': 0.125,
'max_iter': 30
}
for key, val in model_settings:
args[key] = val
intMat, drugMat, targetMat = load_data_from_file(
dataset, os.path.join(data_dir, 'datasets'))
drug_names, target_names = get_drugs_targets_names(
dataset, os.path.join(data_dir, 'datasets'))
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv = intMat.T, targetMat, drugMat, 0
cv_data = cross_validation(X, seeds, cv)
if sp_arg == 0 and predict_num == 0:
if method == 'nrlmf':
cv_eval.nrlmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'kbmf':
cv_eval.kbmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if sp_arg == 1 or predict_num > 0:
tic = time.clock()
if method == 'nrlmf':
model = NRLMF(cfix=args['c'],
K1=args['K1'],
K2=args['K2'],
num_factors=args['r'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
alpha=args['alpha'],
beta=args['beta'],
theta=args['theta'],
max_iter=args['max_iter'])
if method == 'netlaprls':
model = NetLapRLS(gamma_d=args['gamma_d'],
gamma_t=args['gamma_t'],
beta_d=args['beta_t'],
beta_t=args['beta_t'])
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'],
gamma=args['gamma'],
sigma=args['sigma'],
avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=args['T'],
sigma=args['sigma'],
alpha=args['alpha'])
if method == 'kbmf':
model = KBMF(num_factors=args['R'])
if method == 'cmf':
model = CMF(K=args['K'],
lambda_l=args['lambda_l'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
max_iter=args['max_iter'])
cmd = str(model)
if predict_num == 0:
print "Dataset:" + dataset + " CVS:" + str(cvs) + "\n" + cmd
aupr_vec, auc_vec = train(model, cv_data, X, D, T)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
print "auc:%.6f, aupr: %.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f" % (
auc_avg, aupr_avg, auc_conf, aupr_conf, time.clock() - tic)
write_metric_vector_to_file(
auc_vec,
os.path.join(
output_dir,
method + "_auc_cvs" + str(cvs) + "_" + dataset + ".txt"))
write_metric_vector_to_file(
aupr_vec,
os.path.join(
output_dir,
method + "_aupr_cvs" + str(cvs) + "_" + dataset + ".txt"))
elif predict_num > 0:
print "Dataset:" + dataset + "\n" + cmd
seed = 7771 if method == 'cmf' else 22
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
x, y = np.where(intMat == 0)
scores = model.predict_scores(zip(x, y), 5)
ii = np.argsort(scores)[::-1]
predict_pairs = [(drug_names[x[i]], target_names[y[i]], scores[i])
for i in ii[:predict_num]]
new_dti_file = os.path.join(
output_dir, "_".join([method, dataset, "new_dti.txt"]))
novel_prediction_analysis(predict_pairs, new_dti_file,
os.path.join(data_dir, 'biodb'))
def main(argv):
try:
opts, args = getopt.getopt(argv, "m:d:f:c:e:s:o:n:p:g:q:r:l:w", [
"method=", "dataset=", "data-dir=", "cvs=", "external=",
"specify-arg=", "method-opt=", "predict-num=", "scoring=", "gpmi=",
"params=", "output-dir=", "log=", "workdir="
])
except getopt.GetoptError:
sys.exit()
# data_dir = os.path.join(os.path.pardir, 'data')
# output_dir = os.path.join(os.path.pardir, 'output')
method = "nrlmf"
dataset = "nr"
data_dir = '.'
output_dir = '.'
cvs, sp_arg, model_settings, predict_num = 1, 1, [], 0
external = 0
scoring = 'auc'
gpmi = None
params = None
workdir = "./"
logfile = 'job.log'
seeds = [7771, 8367, 22, 1812, 4659]
# seeds = np.random.choice(10000, 5, replace=False)
for opt, arg in opts:
if opt == "--method":
method = arg
if opt == "--dataset":
dataset = arg
if opt == "--data-dir":
data_dir = arg
if opt == "--output-dir":
output_dir = arg
if opt == "--cvs":
cvs = int(arg)
if opt == "--external":
external = int(arg)
if opt == "--specify-arg":
sp_arg = int(arg)
if opt == "--method-opt":
model_settings = [s.split('=') for s in str(arg).split()]
if opt == "--predict-num":
predict_num = int(arg)
if opt == "--scoring":
scoring = str(arg)
if opt == "--gpmi":
gpmi = dict()
for s in str(arg).split():
key, val = s.split('=')
gpmi[key] = float(val)
if opt == "--params":
params = read_params(str(arg))
if opt == "--log":
logfile = str(arg)
if opt == "--workdir":
workdir = str(arg)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
# set logger
logger = logging.getLogger("logger")
logger.setLevel(logging.INFO)
filename = logfile
fh = logging.FileHandler(workdir + "/" + filename)
fh.name = filename
logger.addHandler(fh)
# default parameters for each methods
if method == 'nrlmf':
args = {
'c': 5,
'K1': 5,
'K2': 5,
'r': 50,
'lambda_d': 0.125,
'lambda_t': 0.125,
'alpha': 0.25,
'beta': 0.125,
'theta': 0.5,
'max_iter': 100
}
if method == 'nrlmfb':
args = {
'c': 5,
'K1': 5,
'K2': 5,
'r': 50,
'lambda_d': 0.125,
'lambda_t': 0.125,
'alpha': 0.25,
'beta': 0.125,
'theta': 0.5,
'max_iter': 100
}
if method == 'netlaprls':
args = {'gamma_d': 10, 'gamma_t': 10, 'beta_d': 1e-5, 'beta_t': 1e-5}
if method == 'blmnii':
args = {'alpha': 0.7, 'gamma': 1.0, 'sigma': 1.0, 'avg': False}
if method == 'wnngip':
args = {'T': 0.8, 'sigma': 1.0, 'alpha': 0.8}
if method == 'kbmf':
args = {'R': 50}
if method == 'cmf':
args = {
'K': 50,
'lambda_l': 0.5,
'lambda_d': 0.125,
'lambda_t': 0.125,
'max_iter': 30
}
for key, val in model_settings:
args[key] = float(val)
intMat, drugMat, targetMat = load_data_from_file(
dataset, os.path.join(data_dir, 'dataset'))
drug_names, target_names = get_drugs_targets_names(
dataset, os.path.join(data_dir, 'dataset'))
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv = intMat.T, targetMat, drugMat, 0
cv_data = cross_validation(X, seeds, cv)
if cvs == 1: ev_data = external_validation(X, seeds, cv)
if sp_arg == 0 and predict_num == 0 and external == 0:
if method == 'nrlmf':
cv_eval.nrlmf_cv_eval(method,
dataset,
cv_data,
X,
D,
T,
cvs,
args,
logger,
scoring=scoring,
gpmi=gpmi,
params=params)
if method == 'nrlmfb':
cv_eval.nrlmfb_cv_eval(method,
dataset,
cv_data,
X,
D,
T,
cvs,
args,
logger,
scoring=scoring,
gpmi=gpmi,
params=params)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset, cv_data, X, D, T, cvs,
args, logger)
if method == 'kbmf':
cv_eval.kbmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if sp_arg == 0 and predict_num == 0 and external == 1:
if method == 'nrlmf':
ev_eval.nrlmf_ev_eval(method,
ev_data,
X,
D,
T,
logger,
scoring=scoring,
gpmi=gpmi,
params=params)
if method == 'nrlmfb':
ev_eval.nrlmfb_ev_eval(method,
ev_data,
X,
D,
T,
logger,
scoring=scoring,
gpmi=gpmi,
params=params)
if sp_arg == 1 or predict_num > 0:
if method == 'nrlmf':
model = NRLMF(cfix=args['c'],
K1=args['K1'],
K2=args['K2'],
num_factors=args['r'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
alpha=args['alpha'],
beta=args['beta'],
theta=args['theta'],
max_iter=args['max_iter'])
if method == 'nrlmfb':
model = NRLMFb(cfix=args['c'],
K1=args['K1'],
K2=args['K2'],
num_factors=args['r'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
alpha=args['alpha'],
beta=args['beta'],
theta=args['theta'],
max_iter=args['max_iter'],
eta1=args['eta1'],
eta2=args['eta2'])
if method == 'netlaprls':
model = NetLapRLS(gamma_d=args['gamma_d'],
gamma_t=args['gamma_t'],
beta_d=args['beta_t'],
beta_t=args['beta_t'])
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'],
gamma=args['gamma'],
sigma=args['sigma'],
avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=args['T'],
sigma=args['sigma'],
alpha=args['alpha'])
if method == 'kbmf':
model = KBMF(num_factors=args['R'])
if method == 'cmf':
model = CMF(K=args['K'],
lambda_l=args['lambda_l'],
lambda_d=args['lambda_d'],
lambda_t=args['lambda_t'],
max_iter=args['max_iter'])
cmd = str(model)
if predict_num == 0:
tic = time.time()
print("Dataset:" + dataset + " CVS:" + str(cvs) + "\n" + cmd)
aupr_vec, auc_vec = train(model, cv_data, X, D, T)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
print(
"auc:%.6f, aupr:%.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f"
% (auc_avg, aupr_avg, auc_conf, aupr_conf, time.time() - tic))
# write_metric_vector_to_file(auc_vec, os.path.join(output_dir, method+"_auc_cvs"+str(cvs)+"_"+dataset+".txt"))
# write_metric_vector_to_file(aupr_vec, os.path.join(output_dir, method+"_aupr_cvs"+str(cvs)+"_"+dataset+".txt"))
logger.info(
cmd + ', ' +
"auc:%.6f, aupr:%.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f"
% (auc_avg, aupr_avg, auc_conf, aupr_conf, time.time() - tic))
elif predict_num > 0:
print("Dataset:" + dataset + "\n" + cmd)
seed = 7771 if method == 'cmf' else 22
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
x, y = np.where(intMat == 0)
scores = model.predict_scores(zip(x, y), 5)
ii = np.argsort(scores)[::-1]
predict_pairs = [(drug_names[x[i]], target_names[y[i]], scores[i])
for i in ii[:predict_num]]
new_dti_file = os.path.join(
output_dir, "_".join([method, dataset, "new_dti.txt"]))
novel_prediction_analysis(predict_pairs, new_dti_file,
os.path.join(data_dir, 'biodb'))
def main(argv):
try:
opts, args = getopt.getopt(argv, "m:d:f:c:s:o:n:p", ["method=", "dataset=", "data-dir=", "cvs=", "specify-arg=", "method-options=", "predict-num=", "output-dir=", ])
except getopt.GetoptError:
sys.exit()
data_dir = os.path.join(os.path.pardir, 'data')
output_dir = os.path.join(os.path.pardir, 'output')
cvs, sp_arg, model_settings, predict_num = 1, 1, [], 0
seeds = [7771, 8367, 22, 1812, 4659]
# seeds = np.random.choice(10000, 5, replace=False)
for opt, arg in opts:
if opt == "--method":
method = arg
if opt == "--dataset":
dataset = arg
if opt == "--data-dir":
data_dir = arg
if opt == "--output-dir":
output_dir = arg
if opt == "--cvs":
cvs = int(arg)
if opt == "--specify-arg":
sp_arg = int(arg)
if opt == "--method-options":
model_settings = [s.split('=') for s in str(arg).split()]
if opt == "--predict-num":
predict_num = int(arg)
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
# default parameters for each methods
if method == 'nrlmf':
args = {'c': 5, 'K1': 5, 'K2': 5, 'r': 50, 'lambda_d': 0.125, 'lambda_t': 0.125, 'alpha': 0.25, 'beta': 0.125, 'theta': 0.5, 'max_iter': 100}
# args = {'c': 5, 'K1': 2, 'K2': 2, 'r': 3, 'lambda_d': 0.125, 'lambda_t': 0.125, 'alpha': 0.25, 'beta': 0.125, 'theta': 0.5, 'max_iter': 100}
if method == 'netlaprls':
args = {'gamma_d': 10, 'gamma_t': 10, 'beta_d': 1e-5, 'beta_t': 1e-5}
if method == 'blmnii':
args = {'alpha': 0.7, 'gamma': 1.0, 'sigma': 1.0, 'avg': False}
if method == 'wnngip':
args = {'T': 0.8, 'sigma': 1.0, 'alpha': 0.8}
if method == 'kbmf':
args = {'R': 50}
if method == 'cmf':
args = {'K': 50, 'lambda_l': 0.5, 'lambda_d': 0.125, 'lambda_t': 0.125, 'max_iter': 30}
for key, val in model_settings:
args[key] = val
#ZINC test lines
drug_names, target_names = get_drug_target_names_zinc()
tprs=[] #list storing TPR35 for each CV
RCRS=[] #list storing rcrs for each CV #rcrs is a list of lists
for testnum in range(1,11): #for 10-fold CV
intMat, testMat, drugMat, targetMat = load_data_from_file_csv_10cv(testnum, os.path.join(data_dir))
#ZINC test lines
# drug_names, target_names = get_drugs_targets_names(dataset, os.path.join(data_dir))
#Demo lines
# intMat, testMat, drugMat, targetMat = load_data_from_file_demo(dataset, os.path.join(data_dir))
# drug_names, target_names = get_drug_target_names_demo()
#demo lines
if predict_num == 0:
if cvs == 1: # CV setting CVS1
X, D, T, cv = intMat, drugMat, targetMat, 1
if cvs == 2: # CV setting CVS2
X, D, T, cv = intMat, drugMat, targetMat, 0
if cvs == 3: # CV setting CVS3
X, D, T, cv = intMat.T, targetMat, drugMat, 0
cv_data = cross_validation(X, seeds, cv)
if sp_arg == 0 and predict_num == 0:
if method == 'nrlmf':
cv_eval.nrlmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'netlaprls':
cv_eval.netlaprls_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'blmnii':
cv_eval.blmnii_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'wnngip':
cv_eval.wnngip_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'kbmf':
cv_eval.kbmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if method == 'cmf':
cv_eval.cmf_cv_eval(method, dataset, cv_data, X, D, T, cvs, args)
if sp_arg == 1 or predict_num > 0:
tic = time.clock()
if method == 'nrlmf':
model = NRLMF(cfix=int(args['c']), K1=int(args['K1']), K2=int(args['K2']), num_factors=int(args['r']), lambda_d=float(args['lambda_d']), lambda_t=float(args['lambda_t']), alpha=float(args['alpha']), beta=float(args['beta']), theta=float(args['theta']), max_iter=int(args['max_iter']))
if method == 'netlaprls':
model = NetLapRLS(gamma_d=float(args['gamma_d']), gamma_t=float(args['gamma_t']), beta_d=float(args['beta_t']), beta_t=float(args['beta_t']))
if method == 'blmnii':
model = BLMNII(alpha=args['alpha'], gamma=args['gamma'], sigma=args['sigma'], avg=args['avg'])
if method == 'wnngip':
model = WNNGIP(T=float(args['T']), sigma=float(args['sigma']), alpha=float(args['alpha']))
if method == 'kbmf':
model = KBMF(num_factors=int(args['R']))
if method == 'cmf':
model = CMF(K=int(args['K']), lambda_l=float(args['lambda_l']), lambda_d=float(args['lambda_d']), lambda_t=float(args['lambda_t']), max_iter=int(args['max_iter']))
cmd = str(model)
if predict_num == 0:
print "Dataset:"+dataset+" CVS:"+str(cvs)+"\n"+cmd
aupr_vec, auc_vec = train(model, cv_data, X, D, T)
aupr_avg, aupr_conf = mean_confidence_interval(aupr_vec)
auc_avg, auc_conf = mean_confidence_interval(auc_vec)
print "auc:%.6f, aupr: %.6f, auc_conf:%.6f, aupr_conf:%.6f, Time:%.6f" % (auc_avg, aupr_avg, auc_conf, aupr_conf, time.clock()-tic)
write_metric_vector_to_file(auc_vec, os.path.join(output_dir, method+"_auc_cvs"+str(cvs)+"_"+dataset+".txt"))
write_metric_vector_to_file(aupr_vec, os.path.join(output_dir, method+"_aupr_cvs"+str(cvs)+"_"+dataset+".txt"))
elif predict_num > 0:
# print "Dataset:"+dataset+"\n"+cmd
seed = 7771 if method == 'cmf' else 22
model.fix_model(intMat, intMat, drugMat, targetMat, seed)
# x, y = np.where(intMat == 0)
x, y = np.where(intMat >= 0) #to pick all pairs including train pairs
scores = model.predict_scores(zip(x, y), 5)
# ii = np.argsort(scores)[::-1]
# predict_pairs = [(drug_names[x[i]], target_names[y[i]], scores[i]) for i in ii[:predict_num]]
# print(predict_pairs)
sarr=np.array(scores)
r, c = np.where(testMat > 0)
rcrs=get_rcrs(sarr,zip(r,c))
RCRS=RCRS+rcrs #extending RCRS for each CV
tpr_top35=TPR_by_cutRank(rcrs,35)
tprs.append(tpr_top35)
# tpr35s=TPR_by_cutRank(RCRS,350)
print "Dataset: "+dataset+" Rank: "+str(args['r'])+" Iter: "+str(args['max_iter'])
print "Avg. TPR35: "+str(np.average(tprs))
print "S.E.M. TPR35: "+str(np.std(tprs)/math.sqrt(len(tprs)))
print "TPR35 values:"
print tprs
print "Rank\tTPR"
for rank in range(1,351):
tpr=TPR_by_cutRank(RCRS,rank)
print "%s\t%s"%(str(rank),str(tpr))