def buildInteractions(species,Degrees,inputType,inputDir,outputdir,interactionDirs,Genes=None,
geneSetType=None,PathwayFilter=None,OntologyID=None,directory=None,expressionFile=None,
obligatorySet=None,secondarySet=None,IncludeExpIDs=False):
global degrees
global outputDir
global inputDataType
global obligatoryList ### Add these if connected to anything
global secondaryQueryIDs
global secondDegreeObligatoryCategories ### Add if common to anything in the input - Indicates systems to apply this to
global symbol_hmdb_db; symbol_hmdb_db={}; global hmdb_symbol_db; hmdb_symbol_db={} ### Create an annotation database for HMDB IDs
global FileName
global intNameShort
secondaryQueryIDs = {}
degrees = Degrees
outputDir = outputdir
inputDataType = inputType
obligatoryList = obligatorySet
secondDegreeObligatoryCategories=[]
intNameShort=''
if obligatoryList == None:
obligatoryList=[]
if expressionFile == None:
expressionFile = inputDir ### If it doesn't contain expression values, view as yellow nodes
if secondarySet!= None and (degrees==1 or degrees=='direct'): ### If degrees == 2, this is redundant
### This currently adds alot of predictions - either make more stringent or currently exclude
secondDegreeObligatoryCategories = secondarySet
if PathwayFilter != None:
if len(PathwayFilter)==1:
FileName = PathwayFilter[0]
if isinstance(PathwayFilter, tuple) or isinstance(PathwayFilter, list):
FileName = string.join(list(PathwayFilter),' ')
FileName = string.replace(FileName,':','-')
else:
FileName = PathwayFilter
if len(FileName)>40:
FileName = FileName[:40]
elif OntologyID != None: FileName = OntologyID
elif Genes != None: FileName = Genes
### Import Ensembl-Symbol annotations
getEnsemblGeneData('AltDatabase/ensembl/'+species+'/'+species+'_Ensembl-annotations.txt')
if len(interactionDirs[0]) == 1: interactionDirs = [interactionDirs]
### Import interaction databases indicated in interactionDirs
for i in interactionDirs:
print i
i = export.findFilename(i)
i=string.split(i,'-')[1]
intNameShort+=i[0]
importInteractionData(interactionDirs)
getHMDBData(species) ### overwrite the symbol annotation from any HMDB that comes from a WikiPathway or KEGG pathway that we also include (for consistent official annotation)
input_IDs = getGeneIDs(Genes)
try:
if isinstance(PathwayFilter, tuple):
for pathway in PathwayFilter:
IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,pathway,OntologyID,directory)
for id in IDs:input_IDs[id]=None
else:
input_IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,PathwayFilter,OntologyID,directory)
except Exception: None
if expressionFile == None or len(expressionFile)==0:
expressionFile = exportSelectedIDs(input_IDs) ### create an expression file
elif IncludeExpIDs: ### Prioritize selection of IDs for interactions WITH the primary query set (not among expression input IDs)
secondaryQueryIDs = importqueryResults(species,expressionFile,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs)
sif_file,symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,dir_file)
output_filename = exportGraphImage(species,sif_file,expressionFile)
return output_filename
def buildInteractions(species,Degrees,inputType,inputDir,outputdir,interactionDirs,Genes=None,
geneSetType=None,PathwayFilter=None,OntologyID=None,directory=None,expressionFile=None,
obligatorySet=None,secondarySet=None,IncludeExpIDs=False):
global degrees
global outputDir
global inputDataType
global obligatoryList ### Add these if connected to anything
global secondaryQueryIDs
global secondDegreeObligatoryCategories ### Add if common to anything in the input - Indicates systems to apply this to
global symbol_hmdb_db; symbol_hmdb_db={}; global hmdb_symbol_db; hmdb_symbol_db={} ### Create an annotation database for HMDB IDs
global FileName
global intNameShort
secondaryQueryIDs = {}
degrees = Degrees
outputDir = outputdir
inputDataType = inputType
obligatoryList = obligatorySet
secondDegreeObligatoryCategories=[]
intNameShort=''
if obligatoryList == None:
obligatoryList=[]
if expressionFile == None:
expressionFile = inputDir ### If it doesn't contain expression values, view as yellow nodes
if secondarySet!= None and (degrees==1 or degrees=='direct'): ### If degrees == 2, this is redundant
### This currently adds alot of predictions - either make more stringent or currently exclude
secondDegreeObligatoryCategories = secondarySet
if PathwayFilter != None:
if len(PathwayFilter)==1:
FileName = PathwayFilter[0]
if isinstance(PathwayFilter, tuple) or isinstance(PathwayFilter, list):
FileName = string.join(list(PathwayFilter),' ')
FileName = string.replace(FileName,':','-')
else:
FileName = PathwayFilter
if len(FileName)>40:
FileName = FileName[:40]
elif OntologyID != None: FileName = OntologyID
elif Genes != None: FileName = Genes
### Import Ensembl-Symbol annotations
getEnsemblGeneData('AltDatabase/ensembl/'+species+'/'+species+'_Ensembl-annotations.txt')
if len(interactionDirs[0]) == 1: interactionDirs = [interactionDirs]
### Import interaction databases indicated in interactionDirs
for i in interactionDirs:
print i
i = export.findFilename(i)
i=string.split(i,'-')[1]
intNameShort+=i[0]
importInteractionData(interactionDirs)
getHMDBData(species) ### overwrite the symbol annotation from any HMDB that comes from a WikiPathway or KEGG pathway that we also include (for consistent official annotation)
input_IDs = getGeneIDs(Genes)
try:
if isinstance(PathwayFilter, tuple):
for pathway in PathwayFilter:
IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,pathway,OntologyID,directory)
for id in IDs:input_IDs[id]=None
else:
input_IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,PathwayFilter,OntologyID,directory)
except Exception: None
if expressionFile == None or len(expressionFile)==0:
expressionFile = exportSelectedIDs(input_IDs) ### create an expression file
elif IncludeExpIDs: ### Prioritize selection of IDs for interactions WITH the primary query set (not among expression input IDs)
secondaryQueryIDs = importqueryResults(species,expressionFile,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs)
sif_file,symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,dir_file)
output_filename = exportGraphImage(species,sif_file,expressionFile)
return output_filename
def interactionPermuteTest(species,Degrees,inputType,inputDir,outputdir,interactionDirs,Genes=None,
geneSetType=None,PathwayFilter=None,OntologyID=None,directory=None,expressionFile=None,
obligatorySet=None,secondarySet=None,IncludeExpIDs=False):
global degrees
global outputDir
global inputDataType
global obligatoryList ### Add these if connected to anything
global secondaryQueryIDs
global secondDegreeObligatoryCategories ### Add if common to anything in the input - Indicates systems to apply this to
global symbol_hmdb_db; symbol_hmdb_db={}; global hmdb_symbol_db; hmdb_symbol_db={} ### Create an annotation database for HMDB IDs
global FileName
secondaryQueryIDs = {}
degrees = Degrees
outputDir = outputdir
inputDataType = inputType
obligatoryList = obligatorySet
secondDegreeObligatoryCategories=[]
if obligatoryList == None:
obligatoryList=[]
if expressionFile == None:
expressionFile = inputDir ### If it doesn't contain expression values, view as yellow nodes
if secondarySet!= None and (degrees==1 or degrees=='direct'): ### If degrees == 2, this is redundant
### This currently adds alot of predictions - either make more stringent or currently exclude
secondDegreeObligatoryCategories = secondarySet
if PathwayFilter != None: FileName = PathwayFilter
elif OntologyID != None: FileName = OntologyID
elif Genes != None: FileName = Genes
### Import Ensembl-Symbol annotations
getEnsemblGeneData('AltDatabase/ensembl/'+species+'/'+species+'_Ensembl-annotations.txt')
### Import interaction databases indicated in interactionDirs
importInteractionData(interactionDirs)
getHMDBData(species) ### overwrite the symbol annotation from any HMDB that comes from a WikiPathway or KEGG pathway that we also include (for consistent official annotation)
input_IDs = getGeneIDs(Genes)
try: input_IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,PathwayFilter,OntologyID,directory)
except Exception: None
permutations = 10000; p = 0
secondaryQueryIDs = importqueryResults(species,expressionFile,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs) ### Get the number of unique genes
sif_file, original_symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,dir_file)
#print len(original_symbol_pair_unique)
ensembl_unique = map(lambda x: x, ensembl_symbol_db)
interaction_lengths = []
import random
while p < permutations:
random_inputs = random.sample(ensembl_unique,len(input_IDs))
random_input_db={}
#print len(random_inputs), len(input_IDs); sys.exit()
for i in random_inputs: random_input_db[i]=i
secondaryQueryIDs = importqueryResults(species,random_inputs,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs)
sif_file, symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,inputDir)
#print len(symbol_pair_unique);sys.exit()
interaction_lengths.append(len(symbol_pair_unique))
p+=1
interaction_lengths.sort(); interaction_lengths.reverse()
y = len(original_symbol_pair_unique)
print 'permuted length distribution:',interaction_lengths
print 'original length:',y
k=0
for i in interaction_lengths:
if i>=y: k+=1
print 'p-value:',float(k)/float(permutations)
def interactionPermuteTest(species,Degrees,inputType,inputDir,outputdir,interactionDirs,Genes=None,
geneSetType=None,PathwayFilter=None,OntologyID=None,directory=None,expressionFile=None,
obligatorySet=None,secondarySet=None,IncludeExpIDs=False):
global degrees
global outputDir
global inputDataType
global obligatoryList ### Add these if connected to anything
global secondaryQueryIDs
global secondDegreeObligatoryCategories ### Add if common to anything in the input - Indicates systems to apply this to
global symbol_hmdb_db; symbol_hmdb_db={}; global hmdb_symbol_db; hmdb_symbol_db={} ### Create an annotation database for HMDB IDs
global FileName
secondaryQueryIDs = {}
degrees = Degrees
outputDir = outputdir
inputDataType = inputType
obligatoryList = obligatorySet
secondDegreeObligatoryCategories=[]
if obligatoryList == None:
obligatoryList=[]
if expressionFile == None:
expressionFile = inputDir ### If it doesn't contain expression values, view as yellow nodes
if secondarySet!= None and (degrees==1 or degrees=='direct'): ### If degrees == 2, this is redundant
### This currently adds alot of predictions - either make more stringent or currently exclude
secondDegreeObligatoryCategories = secondarySet
if PathwayFilter != None: FileName = PathwayFilter
elif OntologyID != None: FileName = OntologyID
elif Genes != None: FileName = Genes
### Import Ensembl-Symbol annotations
getEnsemblGeneData('AltDatabase/ensembl/'+species+'/'+species+'_Ensembl-annotations.txt')
### Import interaction databases indicated in interactionDirs
importInteractionData(interactionDirs)
getHMDBData(species) ### overwrite the symbol annotation from any HMDB that comes from a WikiPathway or KEGG pathway that we also include (for consistent official annotation)
input_IDs = getGeneIDs(Genes)
try: input_IDs = gene_associations.simpleGenePathwayImport(species,geneSetType,PathwayFilter,OntologyID,directory)
except Exception: None
permutations = 10000; p = 0
secondaryQueryIDs = importqueryResults(species,expressionFile,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs) ### Get the number of unique genes
sif_file, original_symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,dir_file)
#print len(original_symbol_pair_unique)
ensembl_unique = map(lambda x: x, ensembl_symbol_db)
interaction_lengths = []
import random
while p < permutations:
random_inputs = random.sample(ensembl_unique,len(input_IDs))
random_input_db={}
#print len(random_inputs), len(input_IDs); sys.exit()
for i in random_inputs: random_input_db[i]=i
secondaryQueryIDs = importqueryResults(species,random_inputs,{})[0]
input_IDs,query_interactions,dir_file = importqueryResults(species,inputDir,input_IDs)
sif_file, symbol_pair_unique = associateQueryGenesWithInteractions(input_IDs,query_interactions,inputDir)
#print len(symbol_pair_unique);sys.exit()
interaction_lengths.append(len(symbol_pair_unique))
p+=1
interaction_lengths.sort(); interaction_lengths.reverse()
y = len(original_symbol_pair_unique)
print 'permuted length distribution:',interaction_lengths
print 'original length:',y
k=0
for i in interaction_lengths:
if i>=y: k+=1
print 'p-value:',float(k)/float(permutations)
