def fetchData(selection,saveDir):
cmd = "cpcc get %s"%selection
res = executeCommand(shlex.split(cmd))
lines = res.split("\n")
files = []
#remove first and two last lines
if lines:
del(lines[0])
del(lines[-1])
del(lines[-1])
index = 0
regex =r"(.*)\.(.*)"
for line in lines:
filename = line.strip().strip(",").split("/")[-1]
if filename!="None": #yes this is a none string from the terminal output
m = re.match(regex,filename)
filename = "%s_%s.%s"%(m.group(1),index,m.group(2))
files+=filename
cmd = "cpcc getf %s[%s]"%(selection,index)
res = executeCommand(shlex.split(cmd))
with open("%s/%s"%(saveDir,filename),"w") as f:
f.write(res)
index+=1
return files
#!/usr/bin/env python
import re
import shlex
from lib.md_tools import executeCommand
from lib.util import *
# gets tpr files from a production workflow
cmd = "cpcc cd %s" % getProjectName()
executeCommand(shlex.split(cmd))
cmd = "cpcc get mdrun.in.tpr"
res = executeCommand(shlex.split(cmd))
lines = res.split("\n")
# remove first and two last lines
del (lines[0])
del (lines[-1])
del (lines[-1])
index = 0
regex = r"(.*)\.(.*)"
if not os.path.exists(ProjectDirectories.TPR_DIR):
os.makedirs(ProjectDirectories.TPR_DIR)
# TODO use CpcUtil.fetchData()
#creates a folder named xtc, aggregates the data in a format that is expected for msmbuilder
#for easy backtracking each traj is suffixed with their project name
parser = ArgumentParser()
parser.add_argument("projects",nargs='*',
help="Path to projects that we want to include data from")
parser.add_argument("--trajtype",default='tmd_backbone',
help="Which of the processed trajectories that we wish to use (the subfolders in the analysis directory)")
args = parser.parse_args()
xtcDir = "XTC"
if not os.path.exists(xtcDir):
executeCommand(["mkdir",'-p',xtcDir])
index=0
for projectDir in args.projects:
print projectDir
regex = ".*/(.*)/"
m = re.match(regex,projectDir)
projectName = m.group(1)
path = "%s/analysis/%s/"%(projectDir,args.trajtype)
xtcs = "%s/*xtc"%(path)
trajs = [os.path.basename(f) for f in sorted(glob.glob(xtcs))]
for traj in trajs:
src = "%s/%s"%(path,traj)
#create dir structure for msmbuilder
def buildMDWorkflow(projectName,gromppPath,filePathList,maxCores=24,maxFiles=None,cmdLine=None):
'''
creates a workflow for md simulation using grompp and mdrun function blocks
'''
cmd = "cpcc start %s"%projectName
executeCommand(shlex.split(cmd))
cmd="cpcc import gromacs"
executeCommand(shlex.split(cmd))
cmd="cpcc instance gromacs::grompps grompp"
executeCommand(shlex.split(cmd))
cmd="cpcc instance gromacs::mdruns mdrun"
executeCommand(shlex.split(cmd))
cmd ="cpcc transact"
executeCommand(shlex.split(cmd))
cmd="cpcc connect grompp:out.tpr mdrun:in.tpr"
executeCommand(shlex.split(cmd))
#maximum number of cores to user per simulation. if not set simulations will be tuned
cmd="cpcc set mdrun.in.resources[0].max.cores %s"%maxCores
executeCommand(shlex.split(cmd))
cmd="cpcc setf grompp.in.top[+] topol.top"
executeCommand(shlex.split(cmd))
cmd ="cpcc setf grompp.in.mdp[+] %s"%gromppPath
executeCommand(shlex.split(cmd))
gros = filePathList
# maxFiles = 1
count = 0
for gro in gros:
cmd="cpcc setf grompp.in.conf[+] %s"%gro
executeCommand(shlex.split(cmd))
count+=1
#if we want to limit the number of files to submit in an easy way
if maxFiles and count==maxFiles:
break
if(cmdLine):
cmd = "cpcc set mdrun.in.cmdline_options[+] %s"%cmdLine
executeCommand(shlex.split(cmd))
cmd="cpcc commit"
executeCommand(shlex.split(cmd))
cmd="cpcc activate"
executeCommand(shlex.split(cmd))
def buildMDWorkflowGLIC(projectName,gromppPath,filePathList,maxCores=24,maxFiles=None,cmdLine=None):
'''
creates a workflow for md simulation using grompp and mdrun function blocks
'''
cmd = "cpcc start %s"%projectName
executeCommand(shlex.split(cmd))
cmd="cpcc import gromacs"
executeCommand(shlex.split(cmd))
cmd="cpcc instance gromacs::grompps grompp"
executeCommand(shlex.split(cmd))
cmd="cpcc instance gromacs::mdruns mdrun"
executeCommand(shlex.split(cmd))
cmd ="cpcc transact"
executeCommand(shlex.split(cmd))
cmd="cpcc connect grompp:out.tpr mdrun:in.tpr"
executeCommand(shlex.split(cmd))
#maximum number of cores to user per simulation. if not set simulations will be tuned
cmd="cpcc set mdrun.in.resources[0].max.cores %s"%maxCores
executeCommand(shlex.split(cmd))
cmd="cpcc setf grompp.in.top[+] topol.top"
executeCommand(shlex.split(cmd))
cmd ="cpcc setf grompp.in.mdp[+] %s"%gromppPath
executeCommand(shlex.split(cmd))
#this is a 2d array
cmd ="cpcc setf grompp.in.include[0][0] topol_Protein_chain_A.itp"
#needed for equilibrations
#cmd ="cpcc setf grompp.in.include[+][+] posre_Protein_chain_A.itp"
executeCommand(shlex.split(cmd))
cmd ="cpcc setf grompp.in.include[0][1] %s"%os.path.join(MDToolsDirectories.OTHER,"ffnonbonded.itp")
executeCommand(shlex.split(cmd))
cmd ="cpcc setf grompp.in.include[0][2] %s"%os.path.join(MDToolsDirectories.OTHER,"ffbonded.itp")
executeCommand(shlex.split(cmd))
cmd ="cpcc setf grompp.in.include[0][3] %s"%os.path.join(MDToolsDirectories.POPC_FF_DIR,"popc.itp")
executeCommand(shlex.split(cmd))
gros = filePathList
# maxFiles = 1
count = 0
for gro in gros:
cmd="cpcc setf grompp.in.conf[+] %s"%gro
executeCommand(shlex.split(cmd))
count+=1
#if we want to limit the number of files to submit in an easy way
if maxFiles and count==maxFiles:
break
if(cmdLine):
cmd = "cpcc set mdrun.in.cmdline_options[+] %s"%cmdLine
executeCommand(shlex.split(cmd))
cmd="cpcc commit"
executeCommand(shlex.split(cmd))
cmd="cpcc activate"
executeCommand(shlex.split(cmd))
#!/usr/bin/env python
import shlex
from lib.cpcUtil import CpcUtil
from lib.md_tools import executeCommand
from lib.util import *
# gets trajectory files from a production workflow
# removes the trajectories in the traj dir.
# useful since they already exist in copernicus and we only need them temporarily for creating analysis trajs
cmd = "rm %s/*" % ProjectDirectories.TRAJ_DIR
if executeCommand(shlex.split(cmd)):
print "trajectories removed"
m = re.match(regex,dir)
number = m.group(1)
runs = "%s/_persistence/*run_*/*xtc"%(dir)
xtc = []
for f in glob.glob(runs):
st=os.stat(f)
if st.st_size>0:
xtc.append(f)
if len(xtc)==0:
print "no runs found in %s"%runs
else:
outfile = "%s/traj_%s.xtc"%(ProjectDirectories.TRAJ_DIR,number)
xtc = sorted(xtc)
cmd = ["trjcat" , "-f"] + xtc +["-o",outfile]
executeCommand(cmd)
#do a trjcat here and store it in the tpr dir
# for dir in mdruns:
# #get the mdrun number
# m = re.match(regex,dir)
# number = m.group(1)
# tpr = "%s/_persistence/run_001/topol.tpr"%(dir)
#
#
# tprout = "topol_%s.tpr"%number
# cmd = ["cp" ,tpr,"%s/%s"%(ProjectDirectories.TPR_DIR,tprout)]
# executeCommand(cmd)
