#- Get all metabolic pathways of all E. coli organisms from KEGG.
eColiOrganisms = FEV_KEGG.KEGG.Organism.Group(
searchString='Escherichia coli K-12').organisms
#- For each organism, combine all pathways to the metabolic network, by UNION operation.
organismEcGraphs = []
for organism in eColiOrganisms:
organismPathways = organism.getMetabolicPathways()
organismSubstanceReactionGraph = SubstanceReactionGraph.fromPathway(
organismPathways)
#- Convert this metabolic network into a substance-ecNumber graph.
organismSubstanceGeneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(
organismSubstanceReactionGraph)
organismSubstanceEcGraph = SubstanceEcGraph.fromSubstanceGeneGraph(
organismSubstanceGeneGraph)
organismEcGraphs.append(organismSubstanceEcGraph)
firstGraph = organismEcGraphs.pop(0)
#- Combine all organisms' networks to a consensus network, by INTERSECT operation, leaving only substances and EC numbers that occur in all organisms.
intersectedEcGraph = firstGraph
intersectedEcGraph = intersectedEcGraph.intersection(organismEcGraphs)
#- Print all EC numbers that occur in all organisms.
output = []
for ecNumber in intersectedEcGraph.getECs():
output.append(ecNumber.__str__())
output.sort()
#- Convert each pathway into a substance-ecNumber graph. This is the incomplete metabolic network. ecoReactionGraph = SubstanceReactionGraph.fromPathway(ecoPathways) ssoReactionGraph = SubstanceReactionGraph.fromPathway(ssoPathways) ecoGeneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(ecoReactionGraph) ssoGeneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(ssoReactionGraph) ecoEnzymeGraph = SubstanceEnzymeGraph.fromSubstanceGeneGraph(ecoGeneGraph) ssoEnzymeGraph = SubstanceEnzymeGraph.fromSubstanceGeneGraph(ssoGeneGraph) #- Remove multifunctional enzymes, meaning enzymes associated with more than one EC number. Helps to reduce false gene duplications. ecoEnzymeGraph.removeMultifunctionalEnzymes() ssoEnzymeGraph.removeMultifunctionalEnzymes() ecoEcGraph = SubstanceEcGraph.fromSubstanceEnzymeGraph(ecoEnzymeGraph) ssoEcGraph = SubstanceEcGraph.fromSubstanceEnzymeGraph(ssoEnzymeGraph) #- Combine both species' networks to a consensus network, by INTERSECTION operation. intersectionEcGraph = ecoEcGraph.intersection(ssoEcGraph) #- Create the networks of EC numbers that only occur in eco. onlyEcoEcGraph = ecoEcGraph.difference(intersectionEcGraph, subtractNodes = False) # New steps: output = [] #- For each eco-only EC number, gather any eco gene which encodes an enzyme associated with this EC number. new_ec_numbers = onlyEcoEcGraph.getECs()
def _SubstanceReactionGraph2SubstanceEcGraph(
self, speciesSubstanceReactionGraph: SubstanceReactionGraph
) -> SubstanceEcGraph:
"""
Converts NUKA's substance-reaction graph into a substance-EC graph. Uses pathway information embedded into the graph object.
Parameters
----------
speciesSubstanceReactionGraph : SubstanceReactionGraph
NUKA's substance-reaction graph.
Returns
-------
SubstanceEcGraph
NUKA's substance-EC graph.
Warnings
--------
This function is special to NUKA and **MUST NOT** be used anywhere else!
"""
# shallow-copy old graph to new graph
graph = SubstanceEcGraph(
speciesSubstanceReactionGraph.underlyingRawGraph)
graph.name = 'Substance-EC NUKA'
# create dict of replacements: reaction -> {EC numbers}
replacementDict = dict()
# for each embedded pathway, get list of 'enzyme' entries
for pathway in speciesSubstanceReactionGraph.pathwaySet:
ecEntryList = [
e for e in pathway.entries.values() if e.type == 'enzyme'
]
# for each EC number, get reactions in which it is involved
for ecEntry in ecEntryList:
reactionIDList = ecEntry.reaction.split()
if len(
reactionIDList
) > 0: # filter EC numbers not associated with any reaction
ecNumberList = ecEntry.name.split()
# replace each reaction with its associated EC number
for reactionID in reactionIDList:
reactionName = reactionID.split(':', 1)[1]
reaction = ReactionID(reactionName)
# save associated EC numbers in a set
ecNumberSet = set()
for ecNumberString in ecNumberList:
ecNumber = EcNumber(
ecNumberString.replace('ec:', ''))
ecNumberSet.add(ecNumber)
# update the replacement dict for the current reaction, adding the newly created EC number set
replacementSet = replacementDict.get(reaction, None)
if replacementSet == None or replacementSet.__class__ != set:
replacementSet = set()
replacementSet.update(ecNumberSet)
replacementDict[reaction] = replacementSet
# get list of all reaction edges. Copy edge list to prevent changes in-place, which would NOT work
edgeList = list(graph.getEdges())
# replace reaction edges with EC number edges, using replacement dict
for edge in edgeList:
substrate, product, reaction = edge
# delete old edge
graph.removeEdge(substrate, product, reaction, False)
# add new edges, according to replacement dict
replacementSet = replacementDict[reaction]
for ecNumber in replacementSet:
graph.addEC(substrate, product, ecNumber, False)
if init_verbosity > 0:
print('calculated ' + graph.name)
return graph
#- Download pathway description as KGML.
eco = Organism('eco')
eco00260 = eco.getPathway('00260')
eco01100 = eco.getPathway('01100')
#- Convert to substance-reaction graph.
eco00260_reactionGraph = SubstanceReactionGraph.fromPathway(eco00260)
eco01100_reactionGraph = SubstanceReactionGraph.fromPathway(eco01100)
#- Convert to substance-gene graph
eco00260_geneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(eco00260_reactionGraph)
eco01100_geneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(eco01100_reactionGraph)
#- Convert to substance-ec graph
eco00260_ecGraph = SubstanceEcGraph.fromSubstanceGeneGraph(eco00260_geneGraph)
eco01100_ecGraph = SubstanceEcGraph.fromSubstanceGeneGraph(eco01100_geneGraph)
#- Get set of EC numbers for each graph.
eco00260_ecNumbers = eco00260_ecGraph.getECs()
eco01100_ecNumbers = eco01100_ecGraph.getECs()
#- Calculate difference of EC number sets.
difference_ecNumbers = eco00260_ecNumbers.difference(eco01100_ecNumbers)
#- Print genes (EC number).
output = []
for ecNumber in difference_ecNumbers:
output.append(ecNumber.__str__())
output.sort()
print(str(len(output)) + ' results')
allNonOverviewPathways = eco.getMetabolicPathways(
includeOverviewMaps=False)
#- Convert to substance-reaction graph.
eco01100_reactionGraph = SubstanceReactionGraph.fromPathway(eco01100)
allNonOverviewPathways_reactionGraph = SubstanceReactionGraph.fromPathway(
allNonOverviewPathways)
#- Convert to substance-gene graph.
eco01100_geneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(
eco01100_reactionGraph)
allNonOverviewPathways_geneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(
allNonOverviewPathways_reactionGraph)
#- Convert to substance-ec graph.
eco01100_ecGraph = SubstanceEcGraph.fromSubstanceGeneGraph(
eco01100_geneGraph)
allNonOverviewPathways_ecGraph = SubstanceEcGraph.fromSubstanceGeneGraph(
allNonOverviewPathways_geneGraph)
#- Get set of EC numbers for each graph.
eco01100_ecNumbers = eco01100_ecGraph.getECs()
allNonOverviewPathways_ecNumbers = allNonOverviewPathways_ecGraph.getECs()
#- Calculate difference of EC number sets.
difference_ecNumbers = eco01100_ecNumbers.difference(
allNonOverviewPathways_ecNumbers)
#- Print EC numbers.
output = []
for ecNumber in difference_ecNumbers:
output.append(ecNumber.__str__())
#- Get 00260 pathways of eco and sso from KEGG.
eco = FEV_KEGG.KEGG.Organism.Organism('eco')
sso = FEV_KEGG.KEGG.Organism.Organism('sso')
eco00260 = eco.getPathway('00260')
sso00260 = sso.getPathway('00260')
#- Convert each pathway into a substance-ecNumber graph. This is the incomplete metabolic network.
ecoReactionGraph = SubstanceReactionGraph.fromPathway(eco00260)
ssoReactionGraph = SubstanceReactionGraph.fromPathway(sso00260)
ecoGeneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(ecoReactionGraph)
ssoGeneGraph = SubstanceGeneGraph.fromSubstanceReactionGraph(ssoReactionGraph)
ecoEcGraph = SubstanceEcGraph.fromSubstanceGeneGraph(ecoGeneGraph)
ssoEcGraph = SubstanceEcGraph.fromSubstanceGeneGraph(ssoGeneGraph)
#- Combine both species' networks to a consensus network, by INTERSECTION operation.
intersectionEcGraph = ecoEcGraph.intersection(ssoEcGraph)
#- Create the networks of EC numbers that only occur in 1) eco, and 2) sso.
onlyEcoEcGraph = ecoEcGraph.difference(intersectionEcGraph, subtractNodes = False)
onlySsoEcGraph = ssoEcGraph.difference(intersectionEcGraph, subtractNodes = False)
#- Print those EC numbers.
print('EC numbers occuring only in eco:')
output = []
for ecNumber in onlyEcoEcGraph.getECs():
output.append(ecNumber.__str__())