def initDomainLevelProteins(domainfile):
handle = open(domainfile, 'r')
proteinsA = {}
proteinsB = {}
orthologGroups = {}
groupsStarted = False
ort = None
lineStarts = ['Group', 'Score', 'Boots', '_____']
#header, protein, start, end
for line in handle.readlines():
if groupsStarted:
if line[0:5] not in lineStarts:
hasA = not line.startswith(' ')
splittedLine = line.split()
temp = ort.getBasicProteins(splittedLine)
for p in temp:
p.__class__ = DomainLevelProtein
if hasA:
splittedHeader = Helper.retrieveDomainHeaderInformation(
splittedLine[0])
temp[0].domain = splittedHeader[1]
temp[0].start = int(splittedHeader[2])
temp[0].end = int(splittedHeader[3])
temp[0].header = splittedLine[0]
proteinsA[temp[0].header] = temp[0]
score = float(splittedLine[1].split('%')[0])
ort.inparalogsA[temp[0].header] = score
if not hasA or len(temp) > 1:
splittedHeader = Helper.retrieveDomainHeaderInformation(
splittedLine[-2])
temp[-1].domain = splittedHeader[1]
temp[-1].start = int(splittedHeader[2])
temp[-1].end = int(splittedHeader[3])
temp[-1].header = splittedLine[-2]
proteinsB[temp[-1].header] = temp[-1]
score = float(splittedLine[-1].split('%')[0])
ort.inparalogsB[temp[-1].header] = score
elif line.startswith('Group'):
ort = OrthologyGroup.getBasicOrthologyGroup(
line, False, orthologGroups)
elif line.startswith('Bootstrap'):
ort.addSeeds(line)
else:
if line.startswith('_'):
groupsStarted = True
handle.close()
return proteinsA, proteinsB, orthologGroups
def initDomainLevelProteins(domainfile):
handle = open(domainfile, 'r')
proteinsA = {}
proteinsB = {}
orthologGroups = {}
groupsStarted = False
ort = None
lineStarts = ['Group', 'Score', 'Boots', '_____']
#header, protein, start, end
for line in handle.readlines():
if groupsStarted:
if line[0:5] not in lineStarts:
hasA = not line.startswith(' ')
splittedLine = line.split()
temp = ort.getBasicProteins(splittedLine)
for p in temp:
p.__class__ = DomainLevelProtein
if hasA:
splittedHeader = Helper.retrieveDomainHeaderInformation(splittedLine[0])
temp[0].domain = splittedHeader[1]
temp[0].start = int(splittedHeader[2])
temp[0].end = int(splittedHeader[3])
temp[0].header = splittedLine[0]
proteinsA[temp[0].header] = temp[0]
score = float(splittedLine[1].split('%')[0])
ort.inparalogsA[temp[0].header] = score
if not hasA or len(temp) > 1:
splittedHeader = Helper.retrieveDomainHeaderInformation(splittedLine[-2])
temp[-1].domain = splittedHeader[1]
temp[-1].start = int(splittedHeader[2])
temp[-1].end = int(splittedHeader[3])
temp[-1].header = splittedLine[-2]
proteinsB[temp[-1].header] = temp[-1]
score = float(splittedLine[-1].split('%')[0])
ort.inparalogsB[temp[-1].header] = score
elif line.startswith('Group'):
ort = OrthologyGroup.getBasicOrthologyGroup(line, False, orthologGroups)
elif line.startswith('Bootstrap'):
ort.addSeeds(line)
else:
if line.startswith('_'):
groupsStarted = True
handle.close()
return proteinsA, proteinsB, orthologGroups
def initGeneLevelProteins(filename, tsvfileA, tsvfileB, useDomains):
proteinsA = {}
proteinsB = {}
orthologGroups = {}
groupsStarted = False
rcp = ConfigParser.RawConfigParser()
rcp.read("orthology.cfg")
cutoff = rcp.getint("Options", "domainlengthcutoff")
if useDomains:
domainsA, shortA = Helper.getDomainsFromTsv(tsvfileA, cutoff)
domainsB, shortB = Helper.getDomainsFromTsv(tsvfileB, cutoff)
handle = open(filename, "r")
ort = None
lineStarts = ["Group", "Score", "Boots", "_____"]
for line in handle.readlines():
if groupsStarted:
if line[0:5] not in lineStarts:
hasA = not line.startswith(" ")
temp = []
splittedLine = line.split()
temp = ort.getBasicProteins(splittedLine)
if hasA:
temp[0].__class__ = GeneLevelProtein
proteinsA[temp[0].accession] = temp[0]
if useDomains:
temp[0].domains = domainsA[temp[0].accession]
score = float(splittedLine[1].split("%")[0])
ort.inparalogsA[temp[0].accession] = score
if not hasA or len(temp) > 1:
temp[-1].__class__ = GeneLevelProtein
proteinsB[temp[-1].accession] = temp[-1]
if useDomains:
temp[-1].domains = domainsB[temp[-1].accession]
score = float(splittedLine[-1].split("%")[0])
ort.inparalogsB[temp[-1].accession] = score
elif line.startswith("Group"):
ort = OrthologyGroup.getBasicOrthologyGroup(line, True, orthologGroups)
elif line.startswith("Bootstrap"):
ort.addSeeds(line)
else:
if line.startswith("_"):
groupsStarted = True
pairsCount = 0
for g in orthologGroups:
pairsCount += len(orthologGroups[g].inparalogsA) * len(orthologGroups[g].inparalogsB)
print pairsCount, "should be the amount of pairs"
print len(orthologGroups), "ortholog groups read from the file"
handle.close()
if useDomains:
return proteinsA, proteinsB, orthologGroups, shortA, shortB
else:
return proteinsA, proteinsB, orthologGroups
