def storeMccTotalEner():
import sys
lst_fn = sys.argv[1]
complx_lst = getLst(lst_fn)
for complx in complx_lst:
extractMccTotalEner(complx)
def lstCmp(path1, path2):
list1 = getLst(path1)
list2 = getLst(path2)
common = []
unique1 = []
unique2 = []
for item in list1:
if item in list2:
common.append(item)
else:
unique1.append(item)
for item in list2:
if not item in list1:
unique2.append(item)
return common, unique1, unique2
def subZ_ScoreSvr_tasks():
"""submit the loose grid jobs for svr parameter selection using z_score data set"""
import lst_sub
import sys
filename = sys.argv[1]
jobs = lst_sub.getLst(filename)
tasks = Z_Svr_tasks(jobs, task_fn = filename)
tasks.displayJobs()
tasks.subLooseGrid()
def xargs(lstfn, pyscpt):
"""emulate the linux xargs fucntion, use each entry as the argument for the script
Arguments:
- `lstfn`: name of the list file
- `pyscpt`: python scirpt
"""
lst = getLst(lstfn)
os.environ["script"] = pyscpt
for task in lst:
os.environ["task"] = task
os.system("python $script $task")
time.sleep(5)
def storeMccTotalEnerParallelly(lst_fn):
"""
parallel version of storeMccTotalEner using multiprocessing
"""
from multiprocessing import Pool
complx_lst = getLst(lst_fn)
pool = Pool()
pool.map(extractMccTotalEner, complx_lst)
pool.close()
pool.join()
def getIndividualPcc(self):
"""
calculate the PCC of each complexes' pre-cluster data
"""
from lst_sub import getLst
cmplxes = getLst(self.lstfn)
for cmplx in cmplxes:
ener_fn = cmplx + '-A-' + self.ener_extd
mcc_fn = cmplx + '-A-' + self.mcc_extd
ener_dt = pd.read_csv(ener_fn, sep='\s+')
mcc_dt = pd.read_csv(mcc_fn, header=None)
mcc_dt.columns = ['mcc']
dt = pd.concat([ener_dt, mcc_dt], axis=1)
dt = dt.sort(columns=['mcc'])
print "for the complex", cmplx
preClusterAna.calcuPCC(dt)
def constructTotalDt(self):
"""construc the total data frame of all the complexes
"""
from lst_sub import getLst
cmplxes = getLst(self.lstfn)
dts = []
for cmplx in cmplxes:
ener_fn = cmplx + '-A-' + self.ener_extd
mcc_fn = cmplx + '-A-' + self.mcc_extd
ener_dt = pd.read_csv(ener_fn, sep='\s+')
mcc_dt = pd.read_csv(mcc_fn, header=None)
mcc_dt.columns = ['mcc']
dt = pd.concat([ener_dt, mcc_dt], axis=1)
dt = dt.sort(columns=['mcc'])
dts.append(dt)
self.total_dt = pd.concat(dts)
self.total_dt = self.total_dt.sort(columns=['mcc'])
self.total_dt.columns = self.total_columns
def getReps_bk(clustering_rslt_file):
"""
read the clustering results, return the representatives from each cluster
"""
print "clustering result\t", clustering_rslt_file
clustering_rslt = getLst(clustering_rslt_file)
clustering_rslt = [int(i) for i in clustering_rslt]
total_grps = max(clustering_rslt) + 1
total_pts = len(clustering_rslt)
mappings = zip(range(total_pts), clustering_rslt)
belongings = []
for grp_num in range(total_grps):
grp_member = [mapping[0] for mapping in mappings if mapping[1] == grp_num]
belongings.append({str(grp_num): grp_member})
reps = []
for belonging in belongings:
reps.append(random.choice(belonging.values()[0]))
return reps
import time
class svm_tasks:
def __init__(self, tasks, task_fn = ' '):
self.tasks = tasks
self.task_fn = task_fn
def displayJobs(self):
"""print the name of the jobs"""
print "get tasks from ", self.task_fn
for job_name in self.tasks:
print job_name
def subFineGrid(self):
"""submit the fine grid job for svc parameter selection"""
for subset in self.tasks:
time.sleep(3)
os.environ['subset'] = subset
print "submit fine grid job: ", subset
os.system('qsub -v var1=$subset -N $subset fine_grid.pbs')
if __name__ == '__main__':
'''iterate the list of orders'''
import lst_sub
import sys
filename = sys.argv[1]
jobs = lst_sub.getLst(filename)
tasks = svm_tasks(jobs, task_fn = filename)
tasks.displayJobs()
tasks.subFineGrid()
from lst_sub import getLst
import NB_classifier as nb
ener_rows_ifn = 'ener_row_name.txt'
ff = '08ff'
low_condi_dist_fn = ff + '_low_decoy.dist'
high_condi_dist_fn = ff + '_high_decoy.dist'
bayes_dist_ofn = ff + '_bayes.txt'
low_condi_dist = nb.loadCondiDistribution(low_condi_dist_fn)
high_condi_dist = nb.loadCondiDistribution(high_condi_dist_fn)
ener_rows = getLst(ener_rows_ifn)
def convertPdfName(dist_tuple):
"""conver the first letter of the distribution name to upper case
"""
name = dist_tuple[0]
name = name.upper()[0] + name[1:]
return (name, dist_tuple[1])
################################################################################
# converting
low_condi_dist = [convertPdfName(i) for i in low_condi_dist]
high_condi_dist = [convertPdfName(i) for i in high_condi_dist]
################################################################################
high_bayes_dists = [[ener_rows[i], high_condi_dist[i][0], high_condi_dist[i][1][0], high_condi_dist[i][1][1]]
for i in range(len(ener_rows))]
low_bayes_dists = [[ener_rows[i], low_condi_dist[i][0], low_condi_dist[i][1][0], low_condi_dist[i][1][1]]
for i in range(len(ener_rows))]
dset = f[group_path]
dset.attrs["prt_conf"] = prt_conf
dset.attrs["lig_conf"] = lig_conf
dset.create_dataset(str(rep_num), data=dt)
f.flush()
f.close()
if __name__ == "__main__":
################################################################################
# # extract track
from multiprocessing import Pool
lst_fn = "exp_lst"
complx_lst = getLst(lst_fn)
# for complx in complx_lst:
# extractTrack(complx)
pool = Pool()
pool.map(extractTrack, complx_lst)
pool.close()
pool.join()
################################################################################
# load mcc and energy components
# lst_fn = "exp_lst"
# complx_lst = getLst(lst_fn)
from pprint import pprint
import subprocess
from glob import glob
from lst_sub import getLst
lst_fn = 'total_lst'
rows = str(10001)
complxes = getLst(lst_fn)
incomplete_results = []
for complx in complxes:
matxs = glob(complx + '/*.trace.mat')
if len(matxs) == 10:
for matx in matxs:
wc = subprocess.check_output(['wc', '-l', matx])
total_row = wc.split(' ')[0]
if total_row != rows:
incomplete_results.append(complx)
print complx
break
else:
incomplete_results.append(complx)
print complx
pprint(incomplete_results)
import sys
import multi_process
from lst_sub import getLst
complx_lst = sys.argv[1]
complxes = getLst(complx_lst)
cmds = [('python', '/home/jaydy/Workspace/script/Pyscripts/prepare_cluster.py', complx)
for complx in complxes]
for cmd in cmds:
print cmd
multi_process.multi_process(cmds, 8)
"""
f = h5py.File(self.h5_path)
if self.non_centralized_path in f:
del f[self.non_centralized_path]
f.create_group(self.non_centralized_path)
subgroup = f[self.non_centralized_path]
subgroup.create_dataset('low_decoy', data=self.low_dt.values)
subgroup.create_dataset('high_decoy', data=self.high_dt.values)
f.flush()
f.close()
if __name__ == "__main__":
import sys
jobs_fn = sys.argv[1]
complxes = getLst(jobs_fn)
multi_decoys = MultiComplexesDecoy(complxes)
################################################################################
# collect raw data
multi_decoys.cleanH5()
multi_decoys.vstackAll()
################################################################################
# prepare for linear force field
# multi_decoys.processMccEner(centralized=True, normed=True)
# multi_decoys.splitHighLow()
# multi_decoys.dumpLinearMccEner()
# multi_decoys.printMccEner()
################################################################################
def loadClusteringRslts(self):
self.rslts = getLst(self.clust_rslt_fn)
self.rslts = [int(i) for i in self.rslts]