我有大约2.5M文件要使用.py脚本处理。
我使用的是超级计算器，但问题不在于它的功能，它的python进程本身每次都会打开和关闭，并且会耗费时间。
我使用循环来获取我想用脚本转换的文件夹中的每个文件。所以$ {line}指的是一个文件，其中每一行都指的是文件夹的每个文件。
有没有办法在打开.py脚本而不是循环python脚本后处理所有文件？
有我的循环代码：
### LOOP ###
while :
do
pythonsh ${RAMDISK}/script.py -l ${RAMDISK}/${line}.pdb -U '' -A hydrogens
done
exit
python脚本只是一个将.pdb转换为.ddb文件的工具，我从Autodock4附带的AutodockTools中找到了这些文件。
答案 0 :(得分：1)
我修改了脚本，添加了-i命令行选项。
这将允许您指定包含配体文件名（每行一个）的文本文件，并在不重新启动Python的情况下处理它们。
您现在应该可以将其称为
pythonsh ./newscript.py -i ./list_of_files.txt -U '' -A hydrogens
注意：这是未经测试的！它应该按照给定的方式工作，但要非常小心！
#!/usr/bin/env python
#
#
#
# $Header: /opt/cvs/python/packages/share1.5/AutoDockTools/Utilities24/prepare_ligand4.py,v 1.5.4.1 2009/04/15 17:41:57 rhuey Exp $
#
# Modified 2016/02/07
# Hugh Bothwell http://stackoverflow.com/users/33258
# Added -i commandline option to process multiple files
#
import os
from MolKit import Read
from AutoDockTools.MoleculePreparation import AD4LigandPreparation
# initialize command-line parameters
#-l: ligand
ligand_filename = None
#-i: file containing ligand-filenames
ligand_listfile = None
# optional parameters
verbose = None
add_bonds = False
#-A: repairs to make: add bonds and/or hydrogens
repairs = ""
#-C default: add gasteiger charges
charges_to_add = 'gasteiger'
#-p preserve charges on specific atom types
preserve_charge_types=''
#-U: cleanup by merging nphs_lps, nphs, lps
cleanup = "nphs_lps"
#-B named rotatable bond type(s) to allow to rotate
#allowed_bonds = ""
allowed_bonds = "backbone"
#-r root
root = 'auto'
#-o outputfilename
outputfilename = None
#-F check_for_fragments
check_for_fragments = False
#-I bonds_to_inactivate
bonds_to_inactivate = ""
#-Z inactivate_all_torsions
inactivate_all_torsions = False
#-g attach_nonbonded_fragments
attach_nonbonded_fragments = False
#-m mode
mode = 'automatic'
#-d dictionary
dict = None
def process_file(fname):
mols = Read(fname)
if verbose: print 'read ', fname
mol = mols[0]
if len(mols)>1:
if verbose:
print "more than one molecule in file"
#use the one molecule with the most atoms
ctr = 1
for m in mols[1:]:
ctr += 1
if len(m.allAtoms)>len(mol.allAtoms):
mol = m
if verbose:
print "mol set to ", ctr, "th molecule with", len(mol.allAtoms), "atoms"
coord_dict = {}
for a in mol.allAtoms: coord_dict[a] = a.coords
mol.buildBondsByDistance()
if charges_to_add is not None:
preserved = {}
preserved_types = preserve_charge_types.split(',')
for t in preserved_types:
if not len(t): continue
ats = mol.allAtoms.get(lambda x: x.autodock_element==t)
for a in ats:
if a.chargeSet is not None:
preserved[a] = [a.chargeSet, a.charge]
if verbose:
print "setting up LPO with mode=", mode,
print "and outputfilename= ", outputfilename
print "and check_for_fragments=", check_for_fragments
print "and bonds_to_inactivate=", bonds_to_inactivate
LPO = AD4LigandPreparation(mol, mode, repairs, charges_to_add,
cleanup, allowed_bonds, root,
outputfilename=outputfilename,
dict=dict, check_for_fragments=check_for_fragments,
bonds_to_inactivate=bonds_to_inactivate,
inactivate_all_torsions=inactivate_all_torsions,
attach_nonbonded_fragments=attach_nonbonded_fragments)
#do something about atoms with too many bonds (?)
#FIX THIS: could be peptide ligand (???)
# ??use isPeptide to decide chargeSet??
if charges_to_add is not None:
#restore any previous charges
for atom, chargeList in preserved.items():
atom._charges[chargeList[0]] = chargeList[1]
atom.chargeSet = chargeList[0]
if verbose: print "returning ", mol.returnCode
bad_list = []
for a in mol.allAtoms:
if a.coords!=coord_dict[a]: bad_list.append(a)
if len(bad_list):
print len(bad_list), ' atom coordinates changed!'
for a in bad_list:
print a.name, ":", coord_dict[a], ' -> ', a.coords
else:
if verbose: print "No change in atomic coordinates"
if mol.returnCode != 0:
sys.stderr.write(mol.returnMsg + "\n")
# sys.exit(mol.returnCode)
if __name__ == '__main__':
import sys
import getopt
def usage():
"Print helpful, accurate usage statement to stdout."
print "Usage: prepare_ligand4.py -l filename"
print
print " Description of command..."
print " -l ligand_filename (.pdb or .mol2 or .pdbq format)"
print " -i list_of_filenames.txt (.pdb or .mol2 or .pdbq format)"
print " Optional parameters:"
print " [-v] verbose output"
print " [-o pdbqt_filename] (default output filename is ligand_filename_stem + .pdbqt)"
print " [-d] dictionary to write types list and number of active torsions "
print " [-A] type(s) of repairs to make:\n\t\t bonds_hydrogens, bonds, hydrogens (default is to do no repairs)"
print " [-C] do not add charges (default is to add gasteiger charges)"
print " [-p] preserve input charges on atom type, eg -p Zn"
print " (default is not to preserve charges on any specific atom type)"
print " [-U] cleanup type:\n\t\t nphs_lps, nphs, lps, '' (default is 'nphs_lps') "
print " [-B] type(s) of bonds to allow to rotate "
print " (default sets 'backbone' rotatable and 'amide' + 'guanidinium' non-rotatable)"
print " [-R] index for root"
print " [-F] check for and use largest non-bonded fragment (default is not to do this)"
print " [-M] interactive (default is automatic output)"
print " [-I] string of bonds to inactivate composed of "
print " of zero-based atom indices eg 5_13_2_10 "
print " will inactivate atoms[5]-atoms[13] bond "
print " and atoms[2]-atoms[10] bond "
print " (default is not to inactivate any specific bonds)"
print " [-Z] inactivate all active torsions "
print " (default is leave all rotatable active except amide and guanidinium)"
print " [-g] attach all nonbonded fragments "
print " (default is not to do this)"
# process command arguments
try:
opt_list, args = getopt.getopt(sys.argv[1:], 'l:i:vo:d:A:Cp:U:B:R:MFI:Zgh')
except getopt.GetoptError, msg:
print 'prepare_ligand4.py: %s' %msg
usage()
sys.exit(2)
#'l:vo:d:A:CKU:B:R:MFI:Zg'
for o, a in opt_list:
#print "o=", o, " a=", a
if o in ('-l', '--l'):
ligand_filename = a
if verbose: print 'set ligand_filename to ', a
if o in ('-i', '--i'):
ligand_listfile = a
if verbose: print 'set ligand_listfile to ', a
if o in ('-v', '--v'):
verbose = True
if verbose: print 'set verbose to ', True
if o in ('-o', '--o'):
outputfilename = a
if verbose: print 'set outputfilename to ', a
if o in ('-d', '--d'):
dict = a
if verbose: print 'set dict to ', a
if o in ('-A', '--A'):
repairs = a
if verbose: print 'set repairs to ', a
if o in ('-C', '--C'):
charges_to_add = None
if verbose: print 'do not add charges'
if o in ('-p', '--p'):
preserve_charge_types+=a
preserve_charge_types+=','
if verbose: print 'preserve initial charges on ', preserve_charge_types
if o in ('-U', '--U'):
cleanup = a
if verbose: print 'set cleanup to merge ', a
if o in ('-B', '--B'):
allowed_bonds = a
if verbose: print 'allow ', a, 'bonds set to rotate'
if o in ('-R', '--R'):
root = a
if verbose: print 'set root to ', root
if o in ('-F', '--F'):
check_for_fragments = True
if verbose: print 'set check_for_fragments to True'
if o in ('-M', '--M'):
mode = a
if verbose: print 'set mode to ', a
if o in ('-I', '--I'):
bonds_to_inactivate = a
if verbose: print 'set bonds_to_inactivate to ', a
if o in ('-Z', '--Z'):
inactivate_all_torsions = True
if verbose: print 'set inactivate_all_torsions to ', inactivate_all_torsions
if o in ('-g', '--g'):
attach_nonbonded_fragments = True
if verbose: print 'set attach_nonbonded_fragments to ', attach_nonbonded_fragments
if o in ('-h', '--'):
usage()
sys.exit()
if ligand_filename:
process_file(ligand_filename)
elif ligand_listfile:
# Python 2.5 does not support `with`
# with open(ligand_listfile) as inf:
# for fname in inf:
# process_file(fname.rstrip())
inf = open(ligand_listfile)
for fname in inf:
process_file(fname.rstrip())
inf.close()
else:
print 'prepare_ligand4: either -l (ligand filename) or -i (ligand listfile) must be specified.'
usage()
sys.exit()
# To execute this command type:
# prepare_ligand4.py -l pdb_file -v