def main():
Pentose_target_compounds = ['D-Xylulose-5P', 'D-Ribulose-5P', 'D-Ribose-5P']
PP_compounds = \
['D-Xylulose-5P',\
'D-Ribulose-5P',\
'D-Ribose-5P',\
'D-Erythrose-4P',\
'D-Sedoheptulose-7P',\
'D-Fructose-6P',\
'D-Glyceraldehyde-3P']
Glycolysis_compounds = \
['Dihydroxyacetone-3P',\
'D-Glyceraldehyde-3P',\
'Bisphosphoglycerate',\
'3-Phosphoglycerate',\
'2-Phosphoglycerate',\
'Phosphoenolpyruvate',\
'Pyruvate']
TCA_compounds = \
['Oxaloacetate',\
'Citrate',\
'cis-Aconitate',\
'D-Isocitrate',\
'2-Ketoglutarate',\
#succinyl-CoA\
'Succinate',\
'Fumarate',\
'Malate',\
]
Biosynthese_compounds = PP_compounds + Glycolysis_compounds + TCA_compounds
pathway_list = []
# fast test
pathway_list.append(("TEST", ['2-Phosphoglycerate'], ['Bisphosphoglycerate'], None))
# Optimality Modules:
pathway_list.append(("Glucose to Fructose", ['D-glucose-6P'], ['D-fructose-6P'], None))
pathway_list.append(("Fructose to Glucose", ['D-fructose-6P'], ['D-glucose-6P'], None))
pathway_list.append(("oxaloacetate+acetyl-CoA to citrate", ['oxaloacetate + acetyl-CoA'], ['citrate'], None))
pathway_list.append(("cis-aconitate to succinate", ['cis-aconitate'], ['succinate'], None))
pathway_list.append(("D-xylose to D-xylulose-5P", ['D-Xylose'], ['D-Xylulose-5P'], None))
pathway_list.append(("D-arabitol to D-xylulose-5P", ['D-Arabitol'], ['D-Xylulose-5P'], None))
pathway_list.append(("L-arabinose to D-xylulose-5P", ['L-Arabinose'], ['D-Xylulose-5P'], None))
pathway_list.append(("L-xylulose to D-xylulose-5P", ['L-Xylulose'], ['D-Xylulose-5P'], None))
pathway_list.append(("ribitol to D-xylulose-5P", ['Ribitol'], ['D-Xylulose-5P'], None))
pathway_list.append(("D-ribose to D-xylulose-5P", ['D-Ribose'], ['D-Xylulose-5P'], None))
pathway_list.append(("Oxaloacetate to 2-Ketoglutarate", ['Oxaloacetate'], ['2-Ketoglutarate'], None))
pathway_list.append(("Citrate to 2-Ketoglutarate", ['Citrate'], ['2-Ketoglutarate'], None))
pathway_list.append(("Pentose Phosphate", ['D-Xylulose-5P + D-Xylulose-5P + D-Ribose-5P'], ['D-Fructose-6P + D-Fructose-6P + D-Glyceraldehyde-3P'], None))
#pathway_list.append(("Pentose Phosphate", ['D-Xylulose-5P + D-Ribose-5P'], ['D-Glyceraldehyde-3P + D-Sedoheptulose-7P'], None))
pathway_list.append(("D-glucose to D-ribulose-5P", ['D-Glucose'], ['D-Ribulose-5P'], None))
pathway_list.append(("D-glucose to D-fructose-16P", ['D-Glucose'], ['D-Fructose-16P'], None))
pathway_list.append(("D-fructose-6P to GAP+DHAP", ['D-Fructose-6P'], ['D-Glyceraldehyde-3P + Dihydroxyacetone-3P'], None))
pathway_list.append(("GAP to 3-PG", ['D-Glyceraldehyde-3P'], ['3-Phosphoglycerate'], None))
pathway_list.append(("GAP to 2-PG", ['D-Glyceraldehyde-3P'], ['2-Phosphoglycerate'], None))
pathway_list.append(("BPG to 3-PG", ['Bisphosphoglycerate'], ['3-Phosphoglycerate'], None))
pathway_list.append(("BPG to 2-PG", ['Bisphosphoglycerate'], ['2-Phosphoglycerate'], None))
pathway_list.append(("BPG to PEP", ['Bisphosphoglycerate'], ['Phosphoenolpyruvate'], None))
pathway_list.append(("3-PG to PYR", ['3-Phosphoglycerate'], ['Pyruvate'], None))
# Biosynthesis
pathway_list.append(("3-PG to L-Serine", ['3-Phosphoglycerate'], ['L-Serine'], None))
pathway_list.append(("L-Serine to Glycine", ['L-Serine'], ['Glycine'], None))
pathway_list.append(("Pyruvate to L-Alanine", ['Pyruvate'], ['L-Alanine'], None))
pathway_list.append(("Synthesis of L-Serine", Biosynthese_compounds, ['L-Serine'], None))
pathway_list.append(("Synthesis of L-Alanine", Biosynthese_compounds, ['L-Alanine'], None))
pathway_list.append(("Synthesis of Glycine", Biosynthese_compounds, ['Glycine'], None))
pathway_list.append(("Synthesis of L-Aspartate", Biosynthese_compounds, ['L-Aspartate'], None))
pathway_list.append(("Synthesis of L-Glutamate", Biosynthese_compounds, ['L-Glutamate'], None))
# Pentose utilization
pathway_list.append(("L-Arabinose to Pentose Phosphate", ['L-Arabinose'], Pentose_target_compounds, None))
pathway_list.append(("D-Xylose to Pentose Phosphate", ['D-Xylose'], Pentose_target_compounds, None))
pathway_list.append(("D-Ribose to Pentose Phosphate", ['D-Ribose'], Pentose_target_compounds, None))
pathway_list.append(("Ribitol to Pentose Phosphate", ['Ribitol'], Pentose_target_compounds, None))
pathway_list.append(("D-Arabitol to Pentose Phosphate", ['D-Arabitol'], Pentose_target_compounds, None))
# Glycolysis
pathway_list.append(("GAP to PYR", ['D-Glyceraldehyde-3P'], ['Pyruvate'], 'PP'))
pathway_list.append(("GAP to DHAP", ['D-Glyceraldehyde-3P'], ['Dihydroxyacetone-3P'], 'PP'))
pathway_list.append(("GAP to PEP", ['D-Glyceraldehyde-3P + H2O'], ['Phosphoenolpyruvate + H2O'], 'PP'))
pathway_list.append(("GAP to 2-PG", ['D-Glyceraldehyde-3P'], ['2-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC to GAP", ['D-Glucose'], ['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(("GLC to PYR", ['Glucose'], ['Pyruvate + Pyruvate'], 'PP'))
pathway_list.append(("GLC-36P to GAP", ['D-Glucose-36P'], ['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(("GLC-6P to GAP", ['D-Glucose-6P'], ['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(("GLC-6P to BPG", ['D-Glucose-6P'], ['Bisphosphoglycerate + Bisphosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to 3-PG", ['D-Glucose-6P'], ['3-Phosphoglycerate + 3-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to 2-PG", ['D-Glucose-6P'], ['2-Phosphoglycerate + 2-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to PEP", ['D-Glucose-6P'], ['Phosphoenolpyruvate + Phosphoenolpyruvate'], 'PP'))
pathway_list.append(("2-PG to PYR", ['2-Phosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(("3-PG to PYR", ['3-Phosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(("BPG to PYR", ['Bisphosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(("BPG to PEP", ['Bisphosphoglycerate'], ['Phosphoenolpyruvate'], 'PP'))
# Arren's project
pathway_list.append(("Glyoxylate + GAP to Pentose", ['Glyoxylate + D-Glyceraldehyde-3P'], Pentose_target_compounds, None))
pathway_list.append(("Glyoxylate + PYR to Pentose", ['Glyoxylate + Pyruvate'], Pentose_target_compounds, None))
pathway_list.append(("Glycolate + GAP to Pentose", ['Glycolate + D-Glyceraldehyde-3P'], Pentose_target_compounds, None))
pathway_list.append(("Glycolate + PYR to Pentose", ['Glycolate + Pyruvate'], Pentose_target_compounds, None))
pathway_names = [pathway[0] for pathway in pathway_list]
pathway_map = {}
for pathway in pathway_list:
pathway_map[pathway[0]] = pathway[1:]
if (len(sys.argv) > 1):
pathway_name = pathway_names[int(sys.argv[1]) - 1]
else:
pathway_name = TkListSelection(pathway_names, "Choose a pathway:")
if (pathway_name == None):
sys.exit(0)
(substrates, products, pruning_method) = pathway_map[pathway_name]
pathfinder = PathFinder(carbon_only=False, pruning_method=pruning_method, ignore_chirality=True)
print "-"*80 + "\nChosen pathway name is %s\n" % pathway_name + "-"*80
util._mkdir("../results")
util._mkdir("../results/" + pathway_name)
html_writer = HtmlWriter("../results/" + pathway_name + ".html")
html_writer.write("<h1><center>%s</center></h1>\n" % pathway_name)
html_writer.write("<ul>\n")
for i in range(len(substrates)):
for j in range(len(products)):
##(subs, prod) = pathfinder.balance_reaction(substrates[i], products[j])
(subs, prod) = (substrates[i], products[j])
pathway_prefix = "pathway_%d_%d" % (i, j)
util._mkdir("../results/" + pathway_name + "/" + pathway_prefix)
pathfinder.solve_pathway(subs, prod, html_writer, pathway_name, pathway_prefix, max_levels=6, stop_after_first_solution=True)
html_writer.flush()
html_writer.write("</ul>\n")
html_writer.display()
return
def main():
Pentose_target_compounds = [
'D-Xylulose-5P', 'D-Ribulose-5P', 'D-Ribose-5P'
]
PP_compounds = \
['D-Xylulose-5P',\
'D-Ribulose-5P',\
'D-Ribose-5P',\
'D-Erythrose-4P',\
'D-Sedoheptulose-7P',\
'D-Fructose-6P',\
'D-Glyceraldehyde-3P']
Glycolysis_compounds = \
['Dihydroxyacetone-3P',\
'D-Glyceraldehyde-3P',\
'Bisphosphoglycerate',\
'3-Phosphoglycerate',\
'2-Phosphoglycerate',\
'Phosphoenolpyruvate',\
'Pyruvate']
TCA_compounds = \
['Oxaloacetate',\
'Citrate',\
'cis-Aconitate',\
'D-Isocitrate',\
'2-Ketoglutarate',\
#succinyl-CoA\
'Succinate',\
'Fumarate',\
'Malate',\
]
Biosynthese_compounds = PP_compounds + Glycolysis_compounds + TCA_compounds
pathway_list = []
# fast test
pathway_list.append(
("TEST", ['2-Phosphoglycerate'], ['Bisphosphoglycerate'], None))
# Optimality Modules:
pathway_list.append(
("Glucose to Fructose", ['D-glucose-6P'], ['D-fructose-6P'], None))
pathway_list.append(
("Fructose to Glucose", ['D-fructose-6P'], ['D-glucose-6P'], None))
pathway_list.append(("oxaloacetate+acetyl-CoA to citrate",
['oxaloacetate + acetyl-CoA'], ['citrate'], None))
pathway_list.append(
("cis-aconitate to succinate", ['cis-aconitate'], ['succinate'], None))
pathway_list.append(
("D-xylose to D-xylulose-5P", ['D-Xylose'], ['D-Xylulose-5P'], None))
pathway_list.append(("D-arabitol to D-xylulose-5P", ['D-Arabitol'],
['D-Xylulose-5P'], None))
pathway_list.append(("L-arabinose to D-xylulose-5P", ['L-Arabinose'],
['D-Xylulose-5P'], None))
pathway_list.append(("L-xylulose to D-xylulose-5P", ['L-Xylulose'],
['D-Xylulose-5P'], None))
pathway_list.append(
("ribitol to D-xylulose-5P", ['Ribitol'], ['D-Xylulose-5P'], None))
pathway_list.append(
("D-ribose to D-xylulose-5P", ['D-Ribose'], ['D-Xylulose-5P'], None))
pathway_list.append(("Oxaloacetate to 2-Ketoglutarate", ['Oxaloacetate'],
['2-Ketoglutarate'], None))
pathway_list.append(
("Citrate to 2-Ketoglutarate", ['Citrate'], ['2-Ketoglutarate'], None))
pathway_list.append(
("Pentose Phosphate", ['D-Xylulose-5P + D-Xylulose-5P + D-Ribose-5P'],
['D-Fructose-6P + D-Fructose-6P + D-Glyceraldehyde-3P'], None))
#pathway_list.append(("Pentose Phosphate", ['D-Xylulose-5P + D-Ribose-5P'], ['D-Glyceraldehyde-3P + D-Sedoheptulose-7P'], None))
pathway_list.append(
("D-glucose to D-ribulose-5P", ['D-Glucose'], ['D-Ribulose-5P'], None))
pathway_list.append(("D-glucose to D-fructose-16P", ['D-Glucose'],
['D-Fructose-16P'], None))
pathway_list.append(("D-fructose-6P to GAP+DHAP", ['D-Fructose-6P'],
['D-Glyceraldehyde-3P + Dihydroxyacetone-3P'], None))
pathway_list.append(
("GAP to 3-PG", ['D-Glyceraldehyde-3P'], ['3-Phosphoglycerate'], None))
pathway_list.append(
("GAP to 2-PG", ['D-Glyceraldehyde-3P'], ['2-Phosphoglycerate'], None))
pathway_list.append(
("BPG to 3-PG", ['Bisphosphoglycerate'], ['3-Phosphoglycerate'], None))
pathway_list.append(
("BPG to 2-PG", ['Bisphosphoglycerate'], ['2-Phosphoglycerate'], None))
pathway_list.append(
("BPG to PEP", ['Bisphosphoglycerate'], ['Phosphoenolpyruvate'], None))
pathway_list.append(
("3-PG to PYR", ['3-Phosphoglycerate'], ['Pyruvate'], None))
# Biosynthesis
pathway_list.append(
("3-PG to L-Serine", ['3-Phosphoglycerate'], ['L-Serine'], None))
pathway_list.append(
("L-Serine to Glycine", ['L-Serine'], ['Glycine'], None))
pathway_list.append(
("Pyruvate to L-Alanine", ['Pyruvate'], ['L-Alanine'], None))
pathway_list.append(
("Synthesis of L-Serine", Biosynthese_compounds, ['L-Serine'], None))
pathway_list.append(
("Synthesis of L-Alanine", Biosynthese_compounds, ['L-Alanine'], None))
pathway_list.append(
("Synthesis of Glycine", Biosynthese_compounds, ['Glycine'], None))
pathway_list.append(("Synthesis of L-Aspartate", Biosynthese_compounds,
['L-Aspartate'], None))
pathway_list.append(("Synthesis of L-Glutamate", Biosynthese_compounds,
['L-Glutamate'], None))
# Pentose utilization
pathway_list.append(("L-Arabinose to Pentose Phosphate", ['L-Arabinose'],
Pentose_target_compounds, None))
pathway_list.append(("D-Xylose to Pentose Phosphate", ['D-Xylose'],
Pentose_target_compounds, None))
pathway_list.append(("D-Ribose to Pentose Phosphate", ['D-Ribose'],
Pentose_target_compounds, None))
pathway_list.append(("Ribitol to Pentose Phosphate", ['Ribitol'],
Pentose_target_compounds, None))
pathway_list.append(("D-Arabitol to Pentose Phosphate", ['D-Arabitol'],
Pentose_target_compounds, None))
# Glycolysis
pathway_list.append(
("GAP to PYR", ['D-Glyceraldehyde-3P'], ['Pyruvate'], 'PP'))
pathway_list.append(("GAP to DHAP", ['D-Glyceraldehyde-3P'],
['Dihydroxyacetone-3P'], 'PP'))
pathway_list.append(("GAP to PEP", ['D-Glyceraldehyde-3P + H2O'],
['Phosphoenolpyruvate + H2O'], 'PP'))
pathway_list.append(
("GAP to 2-PG", ['D-Glyceraldehyde-3P'], ['2-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC to GAP", ['D-Glucose'],
['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(
("GLC to PYR", ['Glucose'], ['Pyruvate + Pyruvate'], 'PP'))
pathway_list.append(("GLC-36P to GAP", ['D-Glucose-36P'],
['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(("GLC-6P to GAP", ['D-Glucose-6P'],
['D-Glyceraldehyde-3P + D-Glyceraldehyde-3P'], 'PP'))
pathway_list.append(("GLC-6P to BPG", ['D-Glucose-6P'],
['Bisphosphoglycerate + Bisphosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to 3-PG", ['D-Glucose-6P'],
['3-Phosphoglycerate + 3-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to 2-PG", ['D-Glucose-6P'],
['2-Phosphoglycerate + 2-Phosphoglycerate'], 'PP'))
pathway_list.append(("GLC-6P to PEP", ['D-Glucose-6P'],
['Phosphoenolpyruvate + Phosphoenolpyruvate'], 'PP'))
pathway_list.append(
("2-PG to PYR", ['2-Phosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(
("3-PG to PYR", ['3-Phosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(
("BPG to PYR", ['Bisphosphoglycerate'], ['Pyruvate'], 'PP'))
pathway_list.append(
("BPG to PEP", ['Bisphosphoglycerate'], ['Phosphoenolpyruvate'], 'PP'))
# Arren's project
pathway_list.append(
("Glyoxylate + GAP to Pentose", ['Glyoxylate + D-Glyceraldehyde-3P'],
Pentose_target_compounds, None))
pathway_list.append(
("Glyoxylate + PYR to Pentose", ['Glyoxylate + Pyruvate'],
Pentose_target_compounds, None))
pathway_list.append(
("Glycolate + GAP to Pentose", ['Glycolate + D-Glyceraldehyde-3P'],
Pentose_target_compounds, None))
pathway_list.append(
("Glycolate + PYR to Pentose", ['Glycolate + Pyruvate'],
Pentose_target_compounds, None))
pathway_names = [pathway[0] for pathway in pathway_list]
pathway_map = {}
for pathway in pathway_list:
pathway_map[pathway[0]] = pathway[1:]
if (len(sys.argv) > 1):
pathway_name = pathway_names[int(sys.argv[1]) - 1]
else:
pathway_name = TkListSelection(pathway_names, "Choose a pathway:")
if (pathway_name == None):
sys.exit(0)
(substrates, products, pruning_method) = pathway_map[pathway_name]
pathfinder = PathFinder(carbon_only=False,
pruning_method=pruning_method,
ignore_chirality=True)
print "-" * 80 + "\nChosen pathway name is %s\n" % pathway_name + "-" * 80
util._mkdir("../results")
util._mkdir("../results/" + pathway_name)
html_writer = HtmlWriter("../results/" + pathway_name + ".html")
html_writer.write("<h1><center>%s</center></h1>\n" % pathway_name)
html_writer.write("<ul>\n")
for i in range(len(substrates)):
for j in range(len(products)):
##(subs, prod) = pathfinder.balance_reaction(substrates[i], products[j])
(subs, prod) = (substrates[i], products[j])
pathway_prefix = "pathway_%d_%d" % (i, j)
util._mkdir("../results/" + pathway_name + "/" + pathway_prefix)
pathfinder.solve_pathway(subs,
prod,
html_writer,
pathway_name,
pathway_prefix,
max_levels=6,
stop_after_first_solution=True)
html_writer.flush()
html_writer.write("</ul>\n")
html_writer.display()
return
#!/usr/bin/python
import sys
import os
import util
from chemconvert import hash2graph
from html_writer import HtmlWriter
from svg import Scene
html = HtmlWriter("../results/hash_list.html")
util._mkdir("../results/hash_list")
for line in util.parse_text_file(sys.argv[1]):
print line
graph = hash2graph(line)
graph.initialize_pos()
scene = graph.svg(Scene(200, 200, font_size=12))
html.write_svg(scene, "../results/hash_list/" + line)
html.display()
#!/usr/bin/python
import sys
import os
import util
from chemconvert import hash2graph
from html_writer import HtmlWriter
from svg import Scene
html = HtmlWriter("../results/hash_list.html")
util._mkdir("../results/hash_list")
for line in util.parse_text_file(sys.argv[1]):
print line
graph = hash2graph(line)
graph.initialize_pos()
scene = graph.svg(Scene(200, 200, font_size=12))
html.write_svg(scene, "../results/hash_list/" + line)
html.display()
class Pathologic:
def __init__(self, modules_file, experiment_name, max_module_size=6):
self.modules_file = modules_file
self.experiment_name = experiment_name
self.max_module_size = max_module_size
util._mkdir("../log")
self.logfile = open("../log/pathologic_" + self.experiment_name + ".log", "w")
util._mkdir("../results")
util._mkdir("../results/pathologic_" + self.experiment_name)
self.html_writer = HtmlWriter("../results/pathologic_" + self.experiment_name + ".html")
self.html_writer.write("<h1>List of optimal and non-optimal pathways</h1>\n")
self.html_writer.write("<ul>\n")
self.line_counter = 0
self.pathfinder = None
def find_shortest_pathways(self, substrate, product, max_levels, stop_after_first_solution=True):
G_subs = compound2graph(substrate)
G_prod = compound2graph(product)
return self.pathfinder.find_shortest_pathway([G_subs], [G_prod], max_levels=max_levels, stop_after_first_solution=stop_after_first_solution)
def get_distance(self, substrate, product, max_levels):
G_subs = compound2graph(substrate)
G_prod = compound2graph(product)
return self.pathfinder.find_distance([G_subs], [G_prod], max_levels=max_levels)
def get_all_pathways(self, substrate, product, max_levels):
"""return all existing pathways with a length <= max_distance.
"""
(original_compound_map, possible_pathways, min_length) = self.find_shortest_pathways(substrate, product, max_levels=max_levels, stop_after_first_solution=False)
if (possible_pathways == []):
return []
else:
scene_list = self.pathfinder.get_all_possible_scenes(original_compound_map, possible_pathways)
return [scene for (cost, scene) in scene_list] # strip off the cost of each pathway
def get_shortest_pathways(self, substrate, product, max_levels):
"""return -1 if the is no path with length <= max_levels.
otherwise a pair containing a list of all pathways with the minimal length and the minimal length itself
"""
(original_compound_map, possible_pathways, min_length) = self.find_shortest_pathways(substrate, product, max_levels=max_levels, stop_after_first_solution=True)
if (possible_pathways == []):
return ([], -1)
else:
scene_list = self.pathfinder.get_all_possible_scenes(original_compound_map, possible_pathways)
return ([scene for (cost, scene) in scene_list], min_length) # strip off the cost of each pathway
def verify_pathway(self, pathway):
sys.stdout.write("Verifying pathway: %s\n" % str(pathway))
for i in range(len(pathway)-1):
sys.stdout.write(" - checking '%s' -> '%s' ... " % tuple(pathway[i:i+2]))
sys.stdout.flush()
distance = self.get_distance(pathway[i], pathway[i+1], 1)
if (distance == -1):
sys.stdout.write("FAILED (not neighbors)\n")
else:
sys.stdout.write("OK\n")
sys.stdout.flush()
def is_optimal(self, pathway, i, j, draw_scenes=False):
sys.stdout.write(str(pathway[i:j+1]) + " ... ")
wt_distance = j - i
if (wt_distance <= 1): # we have verified that this is optimal already
return True
if (draw_scenes):
# try to find at least one path which is shorter than the wild-type path:
(scenes, dist) = self.get_shortest_pathways(pathway[i], pathway[j], wt_distance - 1)
else:
dist = self.get_distance(pathway[i], pathway[j], wt_distance - 1)
if (dist == -1): # which means no shorter path has been found, hence the WT pathway is one of the shortest
sys.stdout.write(" is optimal!\n")
self.html_writer.write("<li><span style=color:green>%s - OPTIMAL</span></li>\n" % str(pathway[i:j+1]))
self.html_writer.flush()
return True
else: # there is a shorter path than the WT one
sys.stdout.write(" is not optimal!\n")
self.html_writer.write("<li>\n <span style=color:red>%s - NOT OPTIMAL</span><br>\n " % str(pathway[i:j+1]))
if (draw_scenes):
#for s in scenes:
s = scenes[0] # draws only the first scene (otherwise it could be too long)
self.html_writer.write_svg(s, "pathologic_" + self.experiment_name + "/pathway_%d_%d_%d" % (self.line_counter, i, j))
self.html_writer.write("\n</li>\n")
self.html_writer.flush()
return False
def find_modules(self, pathway, draw_scenes=False):
self.verify_pathway(pathway)
i = 0
for j in range(2, len(pathway)):
if (j - i >= self.max_module_size):
sys.stdout.write(str(pathway[i:(j+1)]) + " is too long for me, chopping off the head...\n")
i += 1
# shorten the path from it's head until it is optimal (or too short, i.e. length=1)
while ( (j - i) > 1 and (not self.is_optimal(pathway, i, j, draw_scenes=draw_scenes)) ):
i += 1
def analyze(self, carbon_only=True, use_antimotifs=True, draw_scenes=False):
for line in util.parse_text_file("../rec/" + self.modules_file + ".txt"):
if (line[0] == '@'):
line = line[1:]
self.pathfinder = PathFinder(carbon_only=carbon_only, pruning_method=None, ignore_chirality=False, use_antimotifs=use_antimotifs, outstream=self.logfile)
else:
self.pathfinder = PathFinder(carbon_only=carbon_only, pruning_method=None, ignore_chirality=True, use_antimotifs=use_antimotifs, outstream=self.logfile)
pathway = line.split(';')
self.find_modules(pathway, draw_scenes=draw_scenes)
self.line_counter += 1
def analyze_pairs(self, carbon_only=True, use_antimotifs=True, max_distance=4):
distances = [] # the minimal pathway length between the substrate and the product
alternatives = [] # each value is the number of alternative pathways with the minimal distance
line_counter = 0
for line in util.parse_text_file("../rec/" + self.modules_file + ".txt"):
if (line[0] == '@'):
line = line[1:]
self.pathfinder = PathFinder(carbon_only=carbon_only, pruning_method=None, ignore_chirality=False, use_antimotifs=use_antimotifs, outstream=self.logfile)
else:
self.pathfinder = PathFinder(carbon_only=carbon_only, pruning_method=None, ignore_chirality=True, use_antimotifs=use_antimotifs, outstream=self.logfile)
(subs, prod, max_steps) = line.split(";", 2)
if (max_steps in ['-1','inf','']):
sys.stdout.write(subs + " -(?)-> " + prod)
sys.stdout.flush()
(scenes, dist) = self.get_shortest_pathways(subs, prod, max_distance)
else:
dist = int(max_steps)
sys.stdout.write(subs + " -(%d)-> " % dist + prod)
sys.stdout.flush()
scenes = self.get_all_pathways(subs, prod, dist)
if (dist == -1):
sys.stdout.write(", Distance(L) = inf, N = 0\n")
sys.stdout.flush()
distances.append("inf")
alternatives.append(0)
self.html_writer.write("<li><span style=color:red>%s <-> %s (distance > %d)</span></li>\n" % (subs, prod, self.max_module_size))
self.html_writer.flush()
else:
sys.stdout.write(", Distance(L) = %d, N = %d\n" % (dist, len(scenes)))
sys.stdout.flush()
distances.append(dist)
alternatives.append(len(scenes))
self.html_writer.write("<li><span style=color:green>%s <-> %s (distance = %d)</span></li>\n" % (subs, prod, dist))
for i in range(len(scenes)):
self.html_writer.write("<li>")
self.html_writer.write_svg(scenes[i], "pathologic_" + self.experiment_name + "/pair%d_path%d" % (line_counter,i))
self.html_writer.write("</li>\n")
self.html_writer.flush()
line_counter += 1
result_file = open("../results/" + self.experiment_name + ".txt", "w")
result_file.write(str(distances) + "\n" + str(alternatives) + "\n")
result_file.close()
def display(self):
self.html_writer.write("</ul>\n")
self.html_writer.write(self.pathfinder.reactor.antimotif_summary())
self.html_writer.display()
def __del__(self):
del self.pathfinder
self.logfile.close()
class Pathologic:
def __init__(self, modules_file, experiment_name, max_module_size=6):
self.modules_file = modules_file
self.experiment_name = experiment_name
self.max_module_size = max_module_size
util._mkdir("../log")
self.logfile = open(
"../log/pathologic_" + self.experiment_name + ".log", "w")
util._mkdir("../results")
util._mkdir("../results/pathologic_" + self.experiment_name)
self.html_writer = HtmlWriter("../results/pathologic_" +
self.experiment_name + ".html")
self.html_writer.write(
"<h1>List of optimal and non-optimal pathways</h1>\n")
self.html_writer.write("<ul>\n")
self.line_counter = 0
self.pathfinder = None
def find_shortest_pathways(self,
substrate,
product,
max_levels,
stop_after_first_solution=True):
G_subs = compound2graph(substrate)
G_prod = compound2graph(product)
return self.pathfinder.find_shortest_pathway(
[G_subs], [G_prod],
max_levels=max_levels,
stop_after_first_solution=stop_after_first_solution)
def get_distance(self, substrate, product, max_levels):
G_subs = compound2graph(substrate)
G_prod = compound2graph(product)
return self.pathfinder.find_distance([G_subs], [G_prod],
max_levels=max_levels)
def get_all_pathways(self, substrate, product, max_levels):
"""return all existing pathways with a length <= max_distance.
"""
(original_compound_map, possible_pathways,
min_length) = self.find_shortest_pathways(
substrate,
product,
max_levels=max_levels,
stop_after_first_solution=False)
if (possible_pathways == []):
return []
else:
scene_list = self.pathfinder.get_all_possible_scenes(
original_compound_map, possible_pathways)
return [scene for (cost, scene) in scene_list
] # strip off the cost of each pathway
def get_shortest_pathways(self, substrate, product, max_levels):
"""return -1 if the is no path with length <= max_levels.
otherwise a pair containing a list of all pathways with the minimal length and the minimal length itself
"""
(original_compound_map, possible_pathways,
min_length) = self.find_shortest_pathways(
substrate,
product,
max_levels=max_levels,
stop_after_first_solution=True)
if (possible_pathways == []):
return ([], -1)
else:
scene_list = self.pathfinder.get_all_possible_scenes(
original_compound_map, possible_pathways)
return ([scene for (cost, scene) in scene_list], min_length
) # strip off the cost of each pathway
def verify_pathway(self, pathway):
sys.stdout.write("Verifying pathway: %s\n" % str(pathway))
for i in range(len(pathway) - 1):
sys.stdout.write(" - checking '%s' -> '%s' ... " %
tuple(pathway[i:i + 2]))
sys.stdout.flush()
distance = self.get_distance(pathway[i], pathway[i + 1], 1)
if (distance == -1):
sys.stdout.write("FAILED (not neighbors)\n")
else:
sys.stdout.write("OK\n")
sys.stdout.flush()
def is_optimal(self, pathway, i, j, draw_scenes=False):
sys.stdout.write(str(pathway[i:j + 1]) + " ... ")
wt_distance = j - i
if (wt_distance <= 1): # we have verified that this is optimal already
return True
if (draw_scenes):
# try to find at least one path which is shorter than the wild-type path:
(scenes,
dist) = self.get_shortest_pathways(pathway[i], pathway[j],
wt_distance - 1)
else:
dist = self.get_distance(pathway[i], pathway[j], wt_distance - 1)
if (
dist == -1
): # which means no shorter path has been found, hence the WT pathway is one of the shortest
sys.stdout.write(" is optimal!\n")
self.html_writer.write(
"<li><span style=color:green>%s - OPTIMAL</span></li>\n" %
str(pathway[i:j + 1]))
self.html_writer.flush()
return True
else: # there is a shorter path than the WT one
sys.stdout.write(" is not optimal!\n")
self.html_writer.write(
"<li>\n <span style=color:red>%s - NOT OPTIMAL</span><br>\n "
% str(pathway[i:j + 1]))
if (draw_scenes):
#for s in scenes:
s = scenes[
0] # draws only the first scene (otherwise it could be too long)
self.html_writer.write_svg(
s, "pathologic_" + self.experiment_name +
"/pathway_%d_%d_%d" % (self.line_counter, i, j))
self.html_writer.write("\n</li>\n")
self.html_writer.flush()
return False
def find_modules(self, pathway, draw_scenes=False):
self.verify_pathway(pathway)
i = 0
for j in range(2, len(pathway)):
if (j - i >= self.max_module_size):
sys.stdout.write(
str(pathway[i:(j + 1)]) +
" is too long for me, chopping off the head...\n")
i += 1
# shorten the path from it's head until it is optimal (or too short, i.e. length=1)
while (
(j - i) > 1 and
(not self.is_optimal(pathway, i, j, draw_scenes=draw_scenes))):
i += 1
def analyze(self,
carbon_only=True,
use_antimotifs=True,
draw_scenes=False):
for line in util.parse_text_file("../rec/" + self.modules_file +
".txt"):
if (line[0] == '@'):
line = line[1:]
self.pathfinder = PathFinder(carbon_only=carbon_only,
pruning_method=None,
ignore_chirality=False,
use_antimotifs=use_antimotifs,
outstream=self.logfile)
else:
self.pathfinder = PathFinder(carbon_only=carbon_only,
pruning_method=None,
ignore_chirality=True,
use_antimotifs=use_antimotifs,
outstream=self.logfile)
pathway = line.split(';')
self.find_modules(pathway, draw_scenes=draw_scenes)
self.line_counter += 1
def analyze_pairs(self,
carbon_only=True,
use_antimotifs=True,
max_distance=4):
distances = [
] # the minimal pathway length between the substrate and the product
alternatives = [
] # each value is the number of alternative pathways with the minimal distance
line_counter = 0
for line in util.parse_text_file("../rec/" + self.modules_file +
".txt"):
if (line[0] == '@'):
line = line[1:]
self.pathfinder = PathFinder(carbon_only=carbon_only,
pruning_method=None,
ignore_chirality=False,
use_antimotifs=use_antimotifs,
outstream=self.logfile)
else:
self.pathfinder = PathFinder(carbon_only=carbon_only,
pruning_method=None,
ignore_chirality=True,
use_antimotifs=use_antimotifs,
outstream=self.logfile)
(subs, prod, max_steps) = line.split(";", 2)
if (max_steps in ['-1', 'inf', '']):
sys.stdout.write(subs + " -(?)-> " + prod)
sys.stdout.flush()
(scenes,
dist) = self.get_shortest_pathways(subs, prod, max_distance)
else:
dist = int(max_steps)
sys.stdout.write(subs + " -(%d)-> " % dist + prod)
sys.stdout.flush()
scenes = self.get_all_pathways(subs, prod, dist)
if (dist == -1):
sys.stdout.write(", Distance(L) = inf, N = 0\n")
sys.stdout.flush()
distances.append("inf")
alternatives.append(0)
self.html_writer.write(
"<li><span style=color:red>%s <-> %s (distance > %d)</span></li>\n"
% (subs, prod, self.max_module_size))
self.html_writer.flush()
else:
sys.stdout.write(", Distance(L) = %d, N = %d\n" %
(dist, len(scenes)))
sys.stdout.flush()
distances.append(dist)
alternatives.append(len(scenes))
self.html_writer.write(
"<li><span style=color:green>%s <-> %s (distance = %d)</span></li>\n"
% (subs, prod, dist))
for i in range(len(scenes)):
self.html_writer.write("<li>")
self.html_writer.write_svg(
scenes[i], "pathologic_" + self.experiment_name +
"/pair%d_path%d" % (line_counter, i))
self.html_writer.write("</li>\n")
self.html_writer.flush()
line_counter += 1
result_file = open("../results/" + self.experiment_name + ".txt", "w")
result_file.write(str(distances) + "\n" + str(alternatives) + "\n")
result_file.close()
def display(self):
self.html_writer.write("</ul>\n")
self.html_writer.write(self.pathfinder.reactor.antimotif_summary())
self.html_writer.display()
def __del__(self):
del self.pathfinder
self.logfile.close()
