def main(pdbfile=None):
#
# Calculate the matrix of pKa shifts [number of mutations:min distance from active site]
#
# Get the PDB file
#
print
print 'Construct dpKa matrix'
print
print 'Usage: Design_dist_nummuts <pdbfile> <database file> <method>'
print
#
# Are we running in parallel?
#
import os
mpi_rank = None
try:
import mpi
mpi_rank = mpi.rank
mpi_size = mpi.size
print 'MPI rank', mpi_rank
print 'MPI size', mpi_size
if size == 1:
mpi_rank = None
except:
#
# No MPI
#
pass
#
# start the run
#
import sys, os
if len(sys.argv) < 4:
raise 'Incorrect usage'
if not pdbfile:
pdbfile = sys.argv[1]
if len(sys.argv) > 2:
dir = sys.argv[2]
else:
raise 'You have to provide a dir for the output'
os.chdir(dir)
#
# Get the method
#
method = sys.argv[3]
#
# Are we doing this cheap style?
#
mcstep_factor = 1.00
if sys.argv[-1] == 'cheap':
mcstep_factor = 0.01
#
# Set the file name
#
filename = os.path.join(
dir, 'distance_nummuts__' + os.path.split(pdbfile)[1]) + '_' + method
#
# Keep the cheap results separated from the real results
#
if sys.argv[-1] == 'cheap':
filename = filename + 'cheap'
#
# Print the filename of the dictionary
#
print 'Setting dictionary filename to', filename
DB = dictIO(filename)
print 'I am running in %s' % os.getcwd()
#
# Do we have a completed pKa calc for it?
#
import pKaTool.pKaIO
X = pKaTool.pKaIO.pKaIO(pdbfile)
if not X.assess_status():
print 'You have to run a pKa calculation first'
raise Exception()
#
# OK, got the pKa calc. Read results
#
wt_pKas = X.readpka()
groups = wt_pKas.keys()
groups.sort()
#
# Design target: for all groups with pKa value between 2 and 10: +2 and -2
#
import os
if os.path.isfile(filename):
#
# Load old file
#
results = DB.load()
else:
#
# No, no old restuls
#
results = {}
#
# Add the PDB file
#
fd = open(pdbfile)
lines = fd.readlines()
fd.close()
results['pdbfile'] = lines
#
# Add the full wt pKa values
#
results['wt_full'] = wt_pKas
DB.save(results)
# ---------------------------------------
#
# Delete old files
#
Design_pKa.delete_old_files(pdbfile, 'N', 'Y')
#
# Start looping
#
dist_range = range(1, 26)
num_muts_range = range(1, 21)
count = 0
design_low = 2.0
design_high = 10.0
#
# Count the number of designable groups
#
design_groups = []
for group in groups:
pKaval = wt_pKas[group]['pKa']
if pKaval > design_low and pKaval < design_high:
design_groups.append(group)
tot_count = float(
len(design_groups) * 2 * len(dist_range) * len(num_muts_range)) / 100.0
#
# Set the number of MC steps
#
#
#if len(groups)<60:
pKMCsteps = 200000
#else:
# #print 'We have %d groups, so increasing pKMCsteps' %(len(groups))
# pKMCsteps=200000
pKMCsteps = int(mcstep_factor * float(pKMCsteps))
#
# Start of loop
#
groups.sort()
#
# Are we running MPI?
#
if mpi_rank != None:
fraction = 1.0 / float(mpi_size)
print 'I will be doing %.2f %% of the job' % (100.0 * fraction)
num_groups = len(groups)
print 'Specifically I will be doing groups', groups[
int(fraction * mpi_rank *
num_groups):int(fraction * (mpi_rank + 1) * num_groups)]
print fraction * mpi_rank * num_groups, 'to', fraction * (
mpi_rank + 1) * num_groups
groups = groups[int(fraction * mpi_rank *
num_groups):int(fraction * (mpi_rank + 1) *
num_groups)]
#
# Sleep to desynchronize processes
#
import time
time.sleep(mpi_rank)
print mpi_rank, 'done sleeping!'
import sys
sys.stdout.flush()
#
# Do the calculation
#
for group in groups:
#
# Loop over all groups
#
if not results.has_key(group):
results[group] = {}
#
# Is somebody working on this group?
#
#results=DB.update(results)
if results[group].has_key('locked'):
if results[group]['locked'] == 1:
print '%s is locked' % group
continue
#
# Lock group
#
results[group]['locked'] = 1
#
# Force reinitialisation of instance
#
X = None
#
# Loop over number of mutations and distance cut-off
#
pKaval = wt_pKas[group]['pKa']
if pKaval > design_low and pKaval < design_high:
for design in ['+20.0', 'm20.0']:
if not results[group].has_key(design):
results[group][design] = {}
for min_target_dist in dist_range:
for num_muts in num_muts_range:
if not results[group][design].has_key(num_muts):
results[group][design][num_muts] = {}
#
# Print what we are doing
#
print 'Checking: %20s design: %s, dist: %5.1f, nummuts: %3d %%done %5.2f' % (
group, design, min_target_dist, num_muts,
float(count) / tot_count)
count = count + 1
#
# ..
#
if not results[group][design][num_muts].has_key(
min_target_dist):
results[group][design][num_muts][
min_target_dist] = {}
#
# Get the solutions
#
dpKas, data_added, X = get_solutions_dist_nummuts(
pdbfile, group, design, num_muts,
min_target_dist, results[group][design]
[num_muts][min_target_dist], method, pKMCsteps,
X)
results[group][design][num_muts][
min_target_dist] = dpKas.copy()
pass
results = DB.update(results)
else:
results[group]['pKa out of range'] = 1
#
# Unlock group and merge results
#
del results[group]['locked']
results = DB.update(results)
#
# All done
#
print
print 'All done - normal exit'
print
return
def main(pdbfile=None):
#
# Calculate the matrix of pKa shifts [number of mutations:min distance from active site]
#
# Get the PDB file
#
print
print 'Construct dpKa matrix for a single design criterium'
print
print 'Usage: Design_two_targets <pdbfile> <database file> <design statement>'
print
#
# start the run
#
import sys, os
if len(sys.argv) < 4:
raise 'Incorrect usage'
if not pdbfile:
pdbfile = sys.argv[1]
if len(sys.argv) > 2:
dir = sys.argv[2]
else:
raise 'You have to provide a dir for the output'
os.chdir(dir)
#
# Get the method
#
method = 'MC'
#
# Which design are we doing?
#
design_statement = sys.argv[3]
#
# Set the file name
#
filename = os.path.join(
dir, 'designtwo__' +
os.path.split(pdbfile)[1]) + '_' + method + design_statement
#
# Print the filename of the dictionary
#
print 'Setting dictionary filename to', filename
DB = dictIO(filename)
print 'I am running in %s' % os.getcwd()
#
# Get wild type pKa values
#
import pKaTool.pKaIO
X = pKaTool.pKaIO.pKaIO(pdbfile)
if not X.assess_status():
import os
print 'You have to run a pKa calculation first'
raise Exception()
#
# OK, got the pKa calc. Read results
#
wt_pKas = X.readpka()
#
# See if we have an old database file we should work on
#
import os
if os.path.isfile(filename):
#
# Load old file
#
results = DB.load()
else:
#
# No, no old restuls
#
results = {}
#
# Add the PDB file
#
fd = open(pdbfile)
lines = fd.readlines()
fd.close()
results['pdbfile'] = lines
#
# Add the full wt pKa values
#
results['wt_full'] = wt_pKas
DB.save(results)
# ---------------------------------------
#
# Delete old files
#
import Design_pKa
Design_pKa.delete_old_files(pdbfile, 'N', 'N')
#
# Start looping
#
dist_range = range(1, 26)
num_muts_range = range(1, 21)
count = 0
#
# Set the number of MC steps
#
pKMCsteps = 200000
#
# Start of loop
#
#
# Loop over number of mutations and distance cut-off
#
X = None
if not results.has_key(design_statement):
results[design_statement] = {}
for min_target_dist in dist_range:
for num_muts in num_muts_range:
#
# Make sure the dictionary entries are there
#
if not results[design_statement].has_key(num_muts):
results[design_statement][num_muts] = {}
if not results[design_statement][num_muts].has_key(
min_target_dist):
results[design_statement][num_muts][min_target_dist] = {}
#
# Have we done this one yet?
#
x = 'Checking dist: %5.2f #muts: %2d....' % (
float(min_target_dist), num_muts)
print x,
if results[design_statement][num_muts][min_target_dist] == {}:
print 'not done. Designing solutions:'
#
# Get the parameters
#
defaults = set_parameters(pdbfile=pdbfile,
design_statement=design_statement,
min_dist=min_target_dist,
num_muts=num_muts)
params = {}
for key in defaults.keys():
params[key] = defaults[key][0]
#
# Call the design routine
#
if not X:
X = Design_pKa.Design_pKa(params)
solutions, dpKa_dict = Design_pKa.run_opt(defaults, X)
res = dpKa_dict.keys()
res.sort()
#for r in res:
# print r,dpKa_dict[r]
#
# Store the solutions
#
print 'Found these solutions'
print dpKa_dict.keys()
if dpKa_dict.keys() == []:
results[design_statement][num_muts][min_target_dist] = {
method: {
'None': 'No solutions'
}
}
else:
results[design_statement][num_muts][min_target_dist] = {
method: dpKa_dict.copy()
}
results = DB.update(results)
else:
print 'done.'
#
# All done
#
print
print 'All done - normal exit'
print
return
def main(pdbfile=None):
#
# Calculate the matrix of pKa shifts [number of mutations:min distance from active site]
#
# Get the PDB file
#
print
print 'Construct dpKa matrix'
print
print 'Usage: Design_dist_nummuts <pdbfile> <database file> <method>'
print
#
# Are we running in parallel?
#
import os
mpi_rank=None
try:
import mpi
mpi_rank=mpi.rank
mpi_size=mpi.size
print 'MPI rank',mpi_rank
print 'MPI size',mpi_size
if size==1:
mpi_rank=None
except:
#
# No MPI
#
pass
#
# start the run
#
import sys,os
if len(sys.argv)<4:
raise 'Incorrect usage'
if not pdbfile:
pdbfile=sys.argv[1]
if len(sys.argv)>2:
dir=sys.argv[2]
else:
raise 'You have to provide a dir for the output'
os.chdir(dir)
#
# Get the method
#
method=sys.argv[3]
#
# Are we doing this cheap style?
#
mcstep_factor=1.00
if sys.argv[-1]=='cheap':
mcstep_factor=0.01
#
# Set the file name
#
filename=os.path.join(dir,'distance_nummuts__'+os.path.split(pdbfile)[1])+'_'+method
#
# Keep the cheap results separated from the real results
#
if sys.argv[-1]=='cheap':
filename=filename+'cheap'
#
# Print the filename of the dictionary
#
print 'Setting dictionary filename to',filename
DB=dictIO(filename)
print 'I am running in %s' %os.getcwd()
#
# Do we have a completed pKa calc for it?
#
import pKaTool.pKaIO
X=pKaTool.pKaIO.pKaIO(pdbfile)
if not X.assess_status():
print 'You have to run a pKa calculation first'
raise Exception()
#
# OK, got the pKa calc. Read results
#
wt_pKas=X.readpka()
groups=wt_pKas.keys()
groups.sort()
#
# Design target: for all groups with pKa value between 2 and 10: +2 and -2
#
import os
if os.path.isfile(filename):
#
# Load old file
#
results=DB.load()
else:
#
# No, no old restuls
#
results={}
#
# Add the PDB file
#
fd=open(pdbfile)
lines=fd.readlines()
fd.close()
results['pdbfile']=lines
#
# Add the full wt pKa values
#
results['wt_full']=wt_pKas
DB.save(results)
# ---------------------------------------
#
# Delete old files
#
Design_pKa.delete_old_files(pdbfile,'N','Y')
#
# Start looping
#
dist_range=range(1,26)
num_muts_range=range(1,21)
count=0
design_low=2.0
design_high=10.0
#
# Count the number of designable groups
#
design_groups=[]
for group in groups:
pKaval=wt_pKas[group]['pKa']
if pKaval>design_low and pKaval<design_high:
design_groups.append(group)
tot_count=float(len(design_groups)*2*len(dist_range)*len(num_muts_range))/100.0
#
# Set the number of MC steps
#
#
#if len(groups)<60:
pKMCsteps=200000
#else:
# #print 'We have %d groups, so increasing pKMCsteps' %(len(groups))
# pKMCsteps=200000
pKMCsteps=int(mcstep_factor*float(pKMCsteps))
#
# Start of loop
#
groups.sort()
#
# Are we running MPI?
#
if mpi_rank!=None:
fraction=1.0/float(mpi_size)
print 'I will be doing %.2f %% of the job' %(100.0*fraction)
num_groups=len(groups)
print 'Specifically I will be doing groups',groups[int(fraction*mpi_rank*num_groups):int(fraction*(mpi_rank+1)*num_groups)]
print fraction*mpi_rank*num_groups,'to',fraction*(mpi_rank+1)*num_groups
groups=groups[int(fraction*mpi_rank*num_groups):int(fraction*(mpi_rank+1)*num_groups)]
#
# Sleep to desynchronize processes
#
import time
time.sleep(mpi_rank)
print mpi_rank,'done sleeping!'
import sys
sys.stdout.flush()
#
# Do the calculation
#
for group in groups:
#
# Loop over all groups
#
if not results.has_key(group):
results[group]={}
#
# Is somebody working on this group?
#
#results=DB.update(results)
if results[group].has_key('locked'):
if results[group]['locked']==1:
print '%s is locked' %group
continue
#
# Lock group
#
results[group]['locked']=1
#
# Force reinitialisation of instance
#
X=None
#
# Loop over number of mutations and distance cut-off
#
pKaval=wt_pKas[group]['pKa']
if pKaval>design_low and pKaval<design_high:
for design in ['+20.0','m20.0']:
if not results[group].has_key(design):
results[group][design]={}
for min_target_dist in dist_range:
for num_muts in num_muts_range:
if not results[group][design].has_key(num_muts):
results[group][design][num_muts]={}
#
# Print what we are doing
#
print 'Checking: %20s design: %s, dist: %5.1f, nummuts: %3d %%done %5.2f' %(group,design,
min_target_dist,
num_muts,float(count)/tot_count)
count=count+1
#
# ..
#
if not results[group][design][num_muts].has_key(min_target_dist):
results[group][design][num_muts][min_target_dist]={}
#
# Get the solutions
#
dpKas,data_added,X=get_solutions_dist_nummuts(pdbfile,
group,
design,num_muts,
min_target_dist,
results[group][design][num_muts][min_target_dist],
method,pKMCsteps,X)
results[group][design][num_muts][min_target_dist]=dpKas.copy()
pass
results=DB.update(results)
else:
results[group]['pKa out of range']=1
#
# Unlock group and merge results
#
del results[group]['locked']
results=DB.update(results)
#
# All done
#
print
print 'All done - normal exit'
print
return
def main(pdbfile=None):
#
# Calculate the matrix of pKa shifts [number of mutations:min distance from active site]
#
# Get the PDB file
#
print
print 'Construct dpKa matrix for a single design criterium'
print
print 'Usage: Design_two_targets <pdbfile> <database file> <design statement>'
print
#
# start the run
#
import sys,os
if len(sys.argv)<4:
raise 'Incorrect usage'
if not pdbfile:
pdbfile=sys.argv[1]
if len(sys.argv)>2:
dir=sys.argv[2]
else:
raise 'You have to provide a dir for the output'
os.chdir(dir)
#
# Get the method
#
method='MC'
#
# Which design are we doing?
#
design_statement=sys.argv[3]
#
# Set the file name
#
filename=os.path.join(dir,'designtwo__'+os.path.split(pdbfile)[1])+'_'+method+design_statement
#
# Print the filename of the dictionary
#
print 'Setting dictionary filename to',filename
DB=dictIO(filename)
print 'I am running in %s' %os.getcwd()
#
# Get wild type pKa values
#
import pKaTool.pKaIO
X=pKaTool.pKaIO.pKaIO(pdbfile)
if not X.assess_status():
import os
print 'You have to run a pKa calculation first'
raise Exception()
#
# OK, got the pKa calc. Read results
#
wt_pKas=X.readpka()
#
# See if we have an old database file we should work on
#
import os
if os.path.isfile(filename):
#
# Load old file
#
results=DB.load()
else:
#
# No, no old restuls
#
results={}
#
# Add the PDB file
#
fd=open(pdbfile)
lines=fd.readlines()
fd.close()
results['pdbfile']=lines
#
# Add the full wt pKa values
#
results['wt_full']=wt_pKas
DB.save(results)
# ---------------------------------------
#
# Delete old files
#
import Design_pKa
Design_pKa.delete_old_files(pdbfile,'N','N')
#
# Start looping
#
dist_range=range(1,26)
num_muts_range=range(1,21)
count=0
#
# Set the number of MC steps
#
pKMCsteps=200000
#
# Start of loop
#
#
# Loop over number of mutations and distance cut-off
#
X=None
if not results.has_key(design_statement):
results[design_statement]={}
for min_target_dist in dist_range:
for num_muts in num_muts_range:
#
# Make sure the dictionary entries are there
#
if not results[design_statement].has_key(num_muts):
results[design_statement][num_muts]={}
if not results[design_statement][num_muts].has_key(min_target_dist):
results[design_statement][num_muts][min_target_dist]={}
#
# Have we done this one yet?
#
x='Checking dist: %5.2f #muts: %2d....' %(float(min_target_dist),num_muts)
print x,
if results[design_statement][num_muts][min_target_dist]=={}:
print 'not done. Designing solutions:'
#
# Get the parameters
#
defaults=set_parameters(pdbfile=pdbfile,
design_statement=design_statement,
min_dist=min_target_dist,
num_muts=num_muts)
params={}
for key in defaults.keys():
params[key]=defaults[key][0]
#
# Call the design routine
#
if not X:
X=Design_pKa.Design_pKa(params)
solutions,dpKa_dict=Design_pKa.run_opt(defaults,X)
res=dpKa_dict.keys()
res.sort()
#for r in res:
# print r,dpKa_dict[r]
#
# Store the solutions
#
print 'Found these solutions'
print dpKa_dict.keys()
if dpKa_dict.keys()==[]:
results[design_statement][num_muts][min_target_dist]={method:{'None':'No solutions'}}
else:
results[design_statement][num_muts][min_target_dist]={method:dpKa_dict.copy()}
results=DB.update(results)
else:
print 'done.'
#
# All done
#
print
print 'All done - normal exit'
print
return
