dy-smirks.py

import sys

import daylight
from daylight import Smiles, Smirks

def output(smiin, smirks, smiout, comment):
  print ("%s\t%s\t%s\t%s" % (smiin, smirks, smiout, comment))

def main(fname):
	file = open(fname, 'r')
	for line in file.readlines():
		line = line.strip()
		if not line or line.startswith("#"):
			print(line);
			continue
		parts = line.split('\t')
		smiles = parts[0]
		smirks = parts[1]
		comment = ""
		if len(parts) != 2:
			comment = parts[2]
		
		try:
			mol = Smiles.smilin(smiles)
			if not mol:
				output(smiles, smirks, "<bad smiles>", comment)
				continue
			try:
				trans = Smirks.compile(smirks)
				reactions = trans.xtransform(mol,0)
				if len(reactions) == 0:
					output(smiles, smirks, "<no match>", comment)
				else:
					output(smiles, smirks, reactions[0].cansmiles(1), comment)
			except:
				output(smiles, smirks, "<bad smirks>", comment)
		except:
			output(smiles, smirks, "<bad smiles>", comment)

if __name__ == "__main__":
	if len(sys.argv) != 2:
		sys.stderr.write("usage: dy-smirks.py test.smirks > outname\n");		
		sys.exit(1)
	main(sys.argv[1])


"""
From Andrew's documentation
>>> 
>>> mol = Smiles.smilin("c1ccccc1N(=O)(=O)")
>>> trans = Smirks.compile("[c:1][N:2](=[O:6])=[O:4]>>[c:1][N+:2](=[O:6])[O-:4]")
>>> reactions = trans.transform(mol)
>>> print "ANSWERS:", len(reactions)
ANSWERS: 2
>>> for reaction in reactions:
...     print reaction.cansmiles()
...
USMILES: O=N(=O)c1ccccc1>>[O-][N+](=O)c1ccccc1
USMILES: O=N(=O)c1ccccc1>>[O-][N+](=O)c1ccccc1
>>> for reaction in reactions:
...    print reaction.cansmiles(1)
...
ASMILES: [O:1]=[N:8](=[O:2])[c:9]1[cH:6][cH:4][cH:3][cH:5][cH:7]1>>[O-:1][N+:8](=[O:2])[c:9]1[cH:6][cH:4][cH:3][cH:5][cH:7]1
ASMILES: [O:1]=[N:8](=[O:2])[c:9]1[cH:6][cH:4][cH:3][cH:5][cH:7]1>>[O-:1][N+:8](=[O:2])[c:9]1[cH:6][cH:4][cH:3][cH:5][cH:7]1
>>>
"""