def substanceReactionGraph(self) -> SubstanceReactionGraph:
"""
NUKA's substance-reaction graph.
Returns
-------
SubstanceReactionGraph
Contains all substrates/products and all reactions known to KEGG's metabolic pathways.
Raises
------
HTTPError
If any underlying organism, pathway, or gene does not exist.
URLError
If connection to KEGG fails.
Note
----
This SubstanceReactionGraph can **NOT** be converted into a SubstanceGeneGraph, as the pathways do not contain gene information!
"""
mockOrganism = Organism(
'ec'
) # 'ec' is not an organism abbreviation, but merely desribes that pathways shall contain EC numbers as edges. This returns the full pathways not specific to any species.
pathwaysSet = mockOrganism.getMetabolicPathways(
includeOverviewMaps=False)
substanceReactionGraph = SubstanceGraphs.Conversion.KeggPathwaySet2SubstanceReactionGraph(
pathwaysSet, localVerbosity=0)
substanceReactionGraph.name = 'Substance-Reaction NUKA'
if init_verbosity > 0:
print('calculated ' + substanceReactionGraph.name)
return substanceReactionGraph
def getTransientMatches(
self, relevantOrganisms: Iterable[str]) -> List[TransientMatch]:
"""
Get full transient matches, considering only relevant orthologous organisms.
Considering only relevant organisms is necessary, because a gene can have several thousand orthologs, including ones from organisms completely out of scope, while calculating the E-value for each of those matches is rather slow and involves several downloads.
Parameters
----------
relevantOrganisms : Iterable[str]
Iterable of organism abbreviations, for each organism to be considered relevant.
Returns
-------
List[TransientMatch]
List of transient matches. These include E-values, which are slow to calculate, which is why only `relevantOrganisms` are considered.
This means that only matches found in `self.bestMatches` which come from relevant organisms are actually converted to transient matches.
"""
from FEV_KEGG.KEGG.Organism import Organism
transientMatches = []
# for all Matches
for match in self.bestMatches:
organismAbbreviation = match.foundGeneID.organismAbbreviation
if organismAbbreviation in relevantOrganisms: # Match is relevant
# fetch relevant organism's info
databaseSize = Organism(
organismAbbreviation).getNumberOfGenes()
eValue = SequenceComparison.getExpectationValue(
match.bitScore, self.queryLength, match.length,
databaseSize)
# calculate Transient Match
transientMatches.append(TransientMatch.fromMatch(
match, eValue))
return transientMatches
b2366 dsdA [4.3.1.18] D-serine dehydratase
b3616 tdh [1.1.1.103] threonine 3-dehydrogenase
b3617 kbl [2.3.1.29] glycine C-acetyltransferase
Conclusion
----------
Global map 01100 does not necessarily contain all enzymes for an organism.
"""
from FEV_KEGG.Graph.SubstanceGraphs import SubstanceReactionGraph, SubstanceGeneGraph, SubstanceEnzymeGraph
from FEV_KEGG.KEGG.Organism import Organism
if __name__ == '__main__':
#- Download pathway definition 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-enzyme graph
- Some EC numbers themselves are missing.
The combination of all single pathways has the following shortcomings:
- Some EC numbers themselves are missing. Due to missing "reaction tags" (fixable without 01100) AND due to missing gene entries (not fixable without 01100).
Solution A: Use combined single pathways, ignore missing EC numbers.
Solution B: Use combined single pathways, repair by comparison with overview map 01100, adding missing "reaction tags" and missing gene entries.
"""
from FEV_KEGG.Graph.SubstanceGraphs import SubstanceReactionGraph, SubstanceGeneGraph, SubstanceEcGraph
from FEV_KEGG.KEGG.Organism import Organism
if __name__ == '__main__':
#- Download pathway description as KGML.
eco = Organism('eco')
eco01100 = eco.getPathway('01100')
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)