Skip to content

Commit

Permalink
Replaced MrBump molrep/phaser/refmac modules with custom SIMBAD modules
Browse files Browse the repository at this point in the history
  • Loading branch information
@hlasimpk
hlasimpk committed May 3, 2017
1 parent 4388fce commit 4bd9ed9
Showing 1 changed file with 52 additions and 132 deletions.
184 changes: 52 additions & 132 deletions simbad/util/mr_util.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -7,22 +7,22 @@
import os
import pandas
import sys
import tempfile
import types

from simbad.util import simbad_util
from simbad.util import mtz_util
from simbad.util import anomalous_util
from simbad.util import molrep_util
from simbad.util import mtz_util
from simbad.util import phaser_util
from simbad.util import refmac_util
from simbad.util import simbad_util

# Set up MrBUMP imports
if not "CCP4" in os.environ.keys():
if not "CCP4" in os.environ.keys():
raise RuntimeError('CCP4 not found')
mrbump_incl = os.path.join(os.environ['CCP4'], "share", "mrbump", "include")

sys.path.append(os.path.join(mrbump_incl, 'cluster'))
sys.path.append(os.path.join(mrbump_incl, 'parsers'))

import cluster_run
import parse_molrep
import parse_refmac

Expand Down Expand Up @@ -124,11 +124,11 @@ class MrSubmit(object):
If a solution is found and early_term is set to True, the queued jobs will be terminated.
"""

def __init__(self, mtz, mr_program, refine_program, model_dir, output_dir, early_term=True, enam=False):
def __init__(self, mtz, mr_program, refine_program, model_dir, output_dir, early_term=True, enant=False):
"""Initialise MrSubmit class"""
self.input_file = None
self._early_term = None
self._enam = None
self._enant = None
self._mtz = None
self._mr_program = None
self._output_dir = None
Expand All @@ -147,7 +147,7 @@ def __init__(self, mtz, mr_program, refine_program, model_dir, output_dir, early
self._free = None

self.early_term = early_term
self.enam = enam
self.enant = enant
self.model_dir = os.path.abspath(model_dir)
self.mtz = mtz
self.mr_program = mr_program
Expand All @@ -165,14 +165,14 @@ def early_term(self, early_term):
self._early_term = early_term

@property
def enam(self):
def enant(self):
"""Flag to decide if enantiomorphic spacegroups should be trialled"""
return self._enam
return self._enant

@enam.setter
def enam(self, enam):
"""Set the enam flag to true or false"""
self._enam = enam
@enant.setter
def enant(self, enant):
"""Set the enant flag to true or false"""
self._enant = enant

@property
def mtz(self):
Expand Down Expand Up @@ -255,66 +255,51 @@ def output_dir(self, output_dir):
"""Define the output directory"""
self._output_dir = output_dir

@contextmanager
def suppress_stdout(self):
"""Code to suppress the stdout of a function, used here to silence stdout of MrBUMP modules"""
with open(os.devnull, "w") as devnull:
old_stdout = sys.stdout
sys.stdout = devnull
try:
yield
finally:
sys.stdout = old_stdout

def _run_job(self, model):
"""Function to run MR on each model"""
logger.info("Running MR and refinement on %s", model.pdb_code)

# Generate MR input file
self.MR_setup(model)

# Set up MR input paths
mr_pdbin = os.path.join(self.model_dir, '{0}.pdb'.format(model.pdb_code))
mr_workdir = os.path.join(self.output_dir, model.pdb_code, 'mr', self.mr_program)
mr_logfile = os.path.join(mr_workdir, '{0}_mr.log'.format(model.pdb_code))
mr_pdbout = os.path.join(mr_workdir, '{0}_mr_output.pdb'.format(model.pdb_code))

# Set up refinement input paths
ref_workdir = os.path.join(mr_workdir, 'refine')
ref_hklout = os.path.join(ref_workdir, '{0}_refinement_output.mtz'.format(model.pdb_code))
ref_logfile = os.path.join(ref_workdir, '{0}_ref.log'.format(model.pdb_code))
ref_pdbout = os.path.join(ref_workdir, '{0}_refinement_output.pdb'.format(model.pdb_code))

# Run job
cljob = cluster_run.ClusterJob()

if self.input_file:
# Parse input file
cljob.parse_input(self.input_file)

# Create temp files for the mr and refine keyfiles
mr_key = tempfile.NamedTemporaryFile(delete=False)
# Add phaser key word args to the temp file
if self.mr_program.lower() == 'phaser':
if self.enam:
sgalternative = "ALL"
else:
sgalternative = "HAND"

mr_key.write("#---PHASER COMMAND SCRIPT GENERATED BY SIMBAD---\n")
mr_key.write("MODE MR_AUTO\n")
mr_key.write('ROOT "{0}_mr_output"\n'.format(model.pdb_code))
mr_key.write("#---DEFINE DATA---\n")
mr_key.write("HKLIN {0}\n".format(os.path.abspath(self.mtz)))
mr_key.write("LABIN F={0} SIGF={1}\n".format(self.f, self.sigf))
mr_key.write("SGALTERNATIVE SELECT {0}\n".format(sgalternative))
mr_key.write("#---DEFINE ENSEMBLES---\n")
mr_key.write("ENSEMBLE ensemble1 &\n")
mr_key.write(' PDB "{0}" RMS 0.6\n'.format(os.path.join(self.model_dir,
'{0}.pdb'.format(model.pdb_code))))
mr_key.write("#---DEFINE COMPOSITION---")
mr_key.write("COMPOSITION BY SOLVENT\n")
mr_key.write("COMPOSITION PERCENTAGE {0}\n".format(self.solvent))
mr_key.write("#---SEARCH PARAMETERS---\n")
mr_key.write("SEARCH ENSEMBLE ensemble1 NUMBER 1\n")
mr_key.close()
ref_key = tempfile.NamedTemporaryFile(delete=False)

if self.mr_program and self.refine_program:
with self.suppress_stdout():
cljob.run(self.mr_program, self.refine_program, mr_key.name, refine_keyfile=ref_key.name)

# Remove temp files
os.unlink(mr_key.name)
os.unlink(ref_key.name)
if self.mr_program.upper() == 'MOLREP':
# Set up class with MOLREP input arguments
molrep = molrep_util.Molrep(self.enant, self.mtz, mr_logfile, mr_pdbin, mr_pdbout, self.space_group,
mr_workdir)
# Run MOLREP
molrep.run()

# Set up class with REFMAC input arguments
refmac = refmac_util.Refmac(self.mtz, ref_hklout, ref_logfile, mr_pdbout, ref_pdbout, ref_workdir)
# Run REFMAC
refmac.run()

elif self.mr_program.upper() == 'PHASER':
hklout = os.path.join(mr_workdir, '{0}_mr_output.mtz'.format(model.pdb_code))

# Set up class with PHASER input arguments
phaser = phaser_util.Phaser(self.enant, self.f, self.mtz, hklout, mr_logfile, mr_pdbin, mr_pdbout,
self.sigf, self.solvent, mr_workdir)
# Run PHASER
phaser.run()

# Set up class with REFMAC input arguments
refmac = refmac_util.Refmac(hklout, ref_hklout, ref_logfile, mr_pdbout, ref_pdbout, ref_workdir)
# Run REFMAC
refmac.run()

if self.early_term and self.early_term != "False":
try:
Expand Down Expand Up @@ -362,43 +347,18 @@ def get_mtz_info(self, mtz):

def MR_setup(self, model):
"""Code to generate directories for MR and refinement for a model
and generate an input file containing information about the MR/refine run
Parameters
----------
model : class
Class object containing the PDB code for the input model
self.output_dir : str
Output directory input to :obj: MrSubmit
self.model_dir : str
Model directory input to :obj: MrSubmit
self.mr_program : str
molecular replacement program input to :obj: MrSubmit
self.space_group : str
space group input to :obj: MrSubmit
self.mtz : str
MTZ file input to :obj: MrSubmit
self.enam : bool
Enantimorphic space groups flag input to :obj: MrSubmit
self.f : str
F flag input to :obj: MrSubmit
self.sigf : str
SIGF flag input to :obj: MrSubmit
self.free : str
FREE flag input to :obj: MrSubmit
self.solvent : float
Predicted solvent content input to :obj: MrSubmit
self.resolution : float
Resolution input to :obj: MrSubmit
Returns
-------
file
Output directory for molecular replacement
file
Output directory for refinement
file
Input file used by MrBUMP MR module
"""

try:
Expand All @@ -417,46 +377,6 @@ def MR_setup(self, model):
os.mkdir(os.path.join(self.output_dir, model.pdb_code, 'mr'))
os.mkdir(os.path.join(self.output_dir, model.pdb_code, 'mr', 'phaser'))
os.mkdir(os.path.join(self.output_dir, model.pdb_code, 'mr', 'phaser', 'refine'))

# create input file path
input_file = os.path.join(self.output_dir, model.pdb_code, 'input.txt')
dire = os.path.join(self.output_dir, model.pdb_code)
pdbi = os.path.join(self.model_dir, '{0}.pdb'.format(model.pdb_code))

# Assign variables
hklr = os.path.join(dire, 'mr', self.mr_program, '{0}_refinement_input.mtz'.format(model.pdb_code))
hklo = os.path.join(dire, 'mr', self.mr_program, '{0}_mr_output.1.mtz'.format(model.pdb_code))
pdbo = os.path.join(dire, 'mr', self.mr_program, '{0}_mr_output.1.pdb'.format(model.pdb_code))
mrlo = os.path.join(dire, 'mr', self.mr_program, '{0}_mr.log'.format(model.pdb_code))
refh = os.path.join(dire, 'mr', self.mr_program, 'refine', '{0}_refinement_output.mtz'.format(model.pdb_code))
refp = os.path.join(dire, 'mr', self.mr_program, 'refine', '{0}_refinement_output.pdb'.format(model.pdb_code))
refl = os.path.join(dire, 'mr', self.mr_program, 'refine', '{0}_ref.log'.format(model.pdb_code))

# Write input file
with open(input_file, 'w') as f:
f.write("DIRE {0}\n".format(os.path.abspath(dire)))
f.write("SGIN {0}\n".format(self.space_group))
f.write("HKL1 {0}\n".format(self.mtz))
f.write("PDBO {0}\n".format(pdbo))
f.write("MRLO {0}\n".format(mrlo))
f.write("REFH {0}\n".format(refh))
f.write("REFP {0}\n".format(refp))
f.write("REFL {0}\n".format(refl))
f.write("ENAN {0}\n".format(self.enam))
f.write("FPIN {0}\n".format(self.f))
f.write("SIGF {0}\n".format(self.sigf))
f.write("FREE {0}\n".format(self.free))
f.write("SOLV {0}\n".format(self.solvent))
f.write("RESO {0}\n".format(self.resolution))
f.write("PDBI {0}\n".format(pdbi))
if self.mr_program == "molrep":
f.write("HKLR {0}\n".format(hklr))
elif self.mr_program == "phaser":
f.write("HKLO {0}\n".format(hklo))
f.write("HKLR {0}\n".format(hklo))

self.input_file = input_file

return

def multiprocessing(self, results, time_out=60, nproc=2):
Expand Down

0 comments on commit 4bd9ed9

Please sign in to comment.