""" Define Deep reinforcement learning network """
self.mainQN = SSDnetwork(param.h_size,self.env,"main",param.beta,param.n_step)
self.targetQN = SSDnetwork(param.h_size,self.env,"target",param.alpha,param.n_step)
self.trainables = tf.trainable_variables()
self.copyOps = copyGraphOp(self.trainables)
self.copyOps2 = copyGraphOp2(self.trainables)
self.init = tf.global_variables_initializer()
self.saver = tf.train.Saver()
train_model = model()
init = train_model.init
saver = train_model.saver
seq1 = readseq.readseq('lib/Ecoli_31.txt')
seq2 = readseq.readseq('lib/Ecoli_42.txt')
if not os.path.exists(game_env().path):
os.makedirs(game_env().path)
if np.size(os.listdir(game_env().path)) > 0:
resume = FLAGS.resume
else:
resume = False
seq = readseq.readseqs('lib/HEV.txt')
""" Main training step """
with tf.Session() as sess:
if FLAGS.use_GPU:
self.trainables = tf.trainable_variables()
self.targetOps = updateTargetGraph(self.trainables, self.param.tau)
elif FLAGS.model_name == "SSD":
self.mainQN = SSDnetwork(self.param.h_size,self.env,"main",self.LEARNING_RATE,self.param.n_step)
self.targetQN = SSDnetwork(self.param.h_size,self.env,"target",self.LEARNING_RATE,self.param.n_step)
self.trainables = tf.trainable_variables()
self.targetOps = updateTargetGraph(self.trainables, self.param.tau)
self.init = tf.global_variables_initializer()
self.saver = tf.train.Saver()
train_model = model()
init = train_model.init
saver = train_model.saver
seq1 = readseq.readseq('lib/Ecoli_1.txt')
seq2 = readseq.readseq('lib/Ecoli_2.txt')
if not os.path.exists(train_env.path):
os.makedirs(train_env.path)
if np.size(os.listdir(train_env.path)) > 0:
resume = FLAGS.resume
else:
resume = False
""" Main test step """
with tf.Session() as sess:
if FLAGS.use_GPU:
sess = tf.Session(config=tf.ConfigProto(device_count={'GPU': 0}))
else:
start = time.time()
startdate = time.localtime()
for pairs in [["Homo_sapiens_BRCA", "Mus_musculus_BRCA"],
["Mus_musculus_BRCA", "Rattus_norvegicus_BRCA"],
["Rattus_norvegicus_BRCA", "Homo_sapiens_BRCA"],
["Homo_sapiens_ELK1", "Mus_musculus_ELK1"],
["Mus_musculus_ELK1", "Rattus_norvegicus_ELK1"],
["Rattus_norvegicus_ELK1", "Homo_sapiens_ELK1"],
["Homo_sapiens_CCDC91", "Mus_musculus_CCDC91"],
["Mus_musculus_CCDC91", "Rattus_norvegicus_CCDC91"],
["Rattus_norvegicus_CCDC91", "Homo_sapiens_CCDC91"],
["Homo_sapiens_FOXP2", "Mus_musculus_FOXP2"],
["Mus_musculus_FOXP2", "Rattus_norvegicus_FOXP2"],
["Rattus_norvegicus_FOXP2", "Homo_sapiens_FOXP2"]]:
seq1 = readseq.readseq('lib/Mammal/' + pairs[0] + '.txt')
seq2 = readseq.readseq('lib/Mammal/' + pairs[1] + '.txt')
#print(_)
past = time.time()
agent.set(seq1 + "A", seq2 + "A")
rT1, rT2, processingtime, j = agent.Global(sess)
now = time.time()
#NWresult = np.max(NW.match(alignment.HEVseq[_],alignment.HEVseq[__]))
print(pairs, len(seq1), len(seq2), rT2, "time",
str(processingtime) + "s",
str(now - start) + "s")
filename = "result/" + FLAGS.model_name + "/result%04d%02d%02d%02d%02d%02d_%d_%d.txt" % (
startdate.tm_year, startdate.tm_mon, startdate.tm_mday,
self.mainQN = SSDnetwork(self.param.h_size, self.env, "main",
self.LEARNING_RATE, self.param.n_step)
self.targetQN = SSDnetwork(self.param.h_size, self.env, "target",
self.LEARNING_RATE, self.param.n_step)
self.trainables = tf.trainable_variables()
self.targetOps = updateTargetGraph(self.trainables, self.param.tau)
self.init = tf.global_variables_initializer()
self.saver = tf.train.Saver()
train_model = model()
init = train_model.init
saver = train_model.saver
seq1 = readseq.readseq('lib/Rattus_norvegicus.Rnor_6.0.dna_rm.chromosome.X.fa')
seq2 = readseq.readseq('lib/Mus_musculus.GRCm38.dna_rm.chromosome.X.fa')
if not os.path.exists(train_env.path):
os.makedirs(train_env.path)
if np.size(os.listdir(train_env.path)) > 0:
resume = FLAGS.resume
else:
resume = False
""" Main test step """
with tf.Session() as sess:
if FLAGS.use_GPU:
sess = tf.Session(config=tf.ConfigProto(device_count={'GPU': 0}))
else:
sess = tf.Session(config=tf.ConfigProto(device_count={'CPU': 0}))
import numpy as np
import DQNalign.tool.Bio.conventional as conventional
import DQNalign.tool.util.ReadSeq as readseq
import time
import tensorflow as tf
import DQNalign.flags as flags
FLAGS = tf.app.flags.FLAGS
start = time.time()
startdate = time.localtime()
seq1 = readseq.readseq('lib/Prochlorococcus_1.fna')
seq2 = readseq.readseq('lib/Prochlorococcus_2.fna')
past = time.time()
c = conventional.Clustal(True, seq1, seq2)
match = c.pair_align()
now = time.time()
print("Ecoli test")
print("result", match, "time", str(now - past) + "s", str(now - start) + "s")
filename = "result/Clustal/result%04d%02d%02d%02d%02d%02d_Ecoli.txt" % (
startdate.tm_year, startdate.tm_mon, startdate.tm_mday, startdate.tm_hour,
startdate.tm_min, startdate.tm_sec)
file = open(filename, "a")
file.write(str(match) + " " + str(now - past) + " " + str(now - start) + "\n")
file.close()
filename = "img/Clustal/result%04d%02d%02d%02d%02d%02d_Ecoli" % (