File size: 5,856 Bytes
96272bc | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 | #!/usr/bin/python
from __future__ import print_function
import socket
import sys
import os
import subprocess
script_dir = os.path.dirname(os.path.abspath(__file__))
project_root = os.path.dirname(script_dir)
conf_dir = os.path.join(project_root, 'conf')
inputs_dir = os.path.join(script_dir, 'inputs')
use_multiprocessing = True
if use_multiprocessing:
import multiprocessing
max_cpus = 2 # We might want to not run on the full number of cores, as Rosetta take about 2 Gb of memory per instance
###################################################################################################################################################################
# Important: The variables below are set to values that will make the run complete faster (as a tutorial example), but will not give scientifically valid results.
# Please change them to the "normal" default values before a real run.
###################################################################################################################################################################
#rosetta_scripts_path = os.path.expanduser("~/rosetta/source/bin/rosetta_scripts")
rosetta_scripts_path = os.path.expanduser(
"/public/home/scnb9biwet/jiangqq/flex_ddG_tutorial-master/software/rosetta3.9/main/source/bin/rosetta_scripts.default.linuxgccrelease"
)
nstruct = 3 # Normally 35
max_minimization_iter = 5 # Normally 5000
abs_score_convergence_thresh = 200.0 # Normally 1.0
number_backrub_trials = 10 # Normally 35000
backrub_trajectory_stride = 5 # Can be whatever you want, if you would like to see results from earlier time points in the backrub trajectory. 7000 is a reasonable number, to give you three checkpoints for a 35000 step run, but you could also set it to 35000 for quickest run time (as the final minimization and packing steps will only need to be run one time).
path_to_script = os.path.join(conf_dir, 'ddG-backrub.xml')
residue_to_mutate = ('B', 49, '') # Residue position to perfrom saturation mutatagenesis. Format: (Chain, PDB residue number, insertion code).
if not os.path.isfile(rosetta_scripts_path):
print('ERROR: "rosetta_scripts_path" variable must be set to the location of the "rosetta_scripts" binary executable')
print('This file might look something like: "rosetta_scripts.linuxgccrelease"')
raise Exception('Rosetta scripts missing')
def run_flex_ddg_saturation( name, input_path, input_pdb_path, chains_to_move, mut_aa, nstruct_i ):
output_directory = os.path.join( project_root, 'output_saturation', os.path.join( '%s_%s' % (name, mut_aa), '%02d' % nstruct_i ) )
if not os.path.isdir(output_directory):
os.makedirs(output_directory)
mutation_chain, mutation_resi, mutation_icode = residue_to_mutate
resfile_path = os.path.join( output_directory, 'mutate_%s%d%s_to_%s.resfile' % (mutation_chain, mutation_resi, mutation_icode, mut_aa) )
with open( resfile_path, 'w') as f:
# Header must be NATAA, not NATRO. This resfile is used both to pick the mutated
# position and as the task operation for the mutant PackRotamersMover. With NATRO,
# the mutant branch freezes the 8 A neighbor shell while the wild type branch
# repacks it, which biases every ddG upwards by several REU (a self-mutation to the
# native amino acid comes out at +6 instead of 0).
f.write( 'NATAA\nstart\n%d%s %s PIKAA %s\n' % (mutation_resi, mutation_icode, mutation_chain, mut_aa) )
flex_ddg_args = [
os.path.abspath(rosetta_scripts_path),
"-s %s" % os.path.abspath(input_pdb_path),
'-parser:protocol', os.path.abspath(path_to_script),
'-parser:script_vars',
'chainstomove=' + chains_to_move,
'mutate_resfile_relpath=' + os.path.abspath( resfile_path ),
'number_backrub_trials=%d' % number_backrub_trials,
'max_minimization_iter=%d' % max_minimization_iter,
'abs_score_convergence_thresh=%.1f' % abs_score_convergence_thresh,
'backrub_trajectory_stride=%d' % backrub_trajectory_stride ,
'-restore_talaris_behavior',
'-in:file:fullatom',
'-ignore_unrecognized_res',
'-ignore_zero_occupancy false',
'-ex1',
'-ex2',
]
log_path = os.path.join(output_directory, 'rosetta.out')
print( 'Running Rosetta with args:' )
print( ' '.join(flex_ddg_args) )
print( 'Output logged to:', os.path.abspath(log_path) )
print()
outfile = open(log_path, 'w')
process = subprocess.Popen(flex_ddg_args, stdout=outfile, stderr=subprocess.STDOUT, close_fds = True, cwd = output_directory)
returncode = process.wait()
outfile.close()
if __name__ == '__main__':
mutation_chain, mutation_resi, mutation_icode = residue_to_mutate
cases = []
for nstruct_i in range(1, nstruct + 1 ):
for case_name in os.listdir(inputs_dir):
case_path = os.path.join( inputs_dir, case_name )
for f in os.listdir(case_path):
if f.endswith('.pdb'):
input_pdb_path = os.path.join( case_path, f )
break
with open( os.path.join( case_path, 'chains_to_move.txt' ), 'r' ) as f:
chains_to_move = f.readlines()[0].strip()
for mut_aa in 'ACDEFGHIKLMNPQRSTVWY':
cases.append( ('%s_%s%d%s' % (case_name, mutation_chain, mutation_resi, mutation_icode), case_path, input_pdb_path, chains_to_move, mut_aa, nstruct_i) )
if use_multiprocessing:
pool = multiprocessing.Pool( processes = min(max_cpus, multiprocessing.cpu_count()) )
for args in cases:
if use_multiprocessing:
pool.apply_async( run_flex_ddg_saturation, args = args )
else:
run_flex_ddg_saturation( *args )
if use_multiprocessing:
pool.close()
pool.join()
|