def test_proteomeStatistics():
s = summary.Summary(dataPath + 'annotation.pklz')
s.proteomeStatistics(dataPath + "expression", dataPath + "expression_case")
proteome = [x for x in io.readTable("proteome_features.tsv")]
assert len(proteome) == 20
txs = [
x['Transcript'] for x in proteome if x['Feature_type'] == 'Transcript'
]
assert len(txs) == len(set(txs))
# ENSG00.5 is complete, as it only lacks information of its two aberrant transcripts
assert [x for x in proteome if x['GeneId'] == 'ENSG00.5']
assert len([x for x in proteome if x['Feature_type'] == 'Prosite']) == 3
assert len([x for x in proteome if x['Feature_type'] == 'Pfam']) == 3
# ENST08.1 goes first in the file, and both have same median
assert len([
x for x in proteome
if x['Transcript'] == 'ENST08.1' and x['Feature_type'] != 'Transcript'
]) == 2
assert len([x for x in proteome if x['Transcript'] == 'ENST09.1']) == 0
assert len([
x for x in proteome
if x['Transcript'] == 'ENST18.3' and x['Feature_type'] != 'Transcript'
]) == 1
os.remove('proteome_features.tsv')
def getDomainInteractions(self, ddi):
if self._new and not ddi:
raise SpadaError(
"A file containing the domain-domain interactions must be provided."
)
elif not ddi:
self.logger.info(
"Domain-domain interactions from the provided network will be used."
)
return
self.logger.info("Building isoform-isoform interaction network.")
allDDIs = {
frozenset([x['Pfam1'], x['Pfam2']])
for x in io.readTable(ddi, keys=['Pfam1', 'Pfam2'])
}
gene2tx = io.getGene2Tx(self._txs)
for gene1, gene2 in self._genes._net.edges():
for tx1, tx2 in product(gene2tx.get(gene1, set()),
gene2tx.get(gene2, set())):
possibleDDIs = {
frozenset(x)
for x in product(self._txs[tx1]["Pfam"], self._txs[tx2]
["Pfam"])
}
matches = possibleDDIs & allDDIs
if matches:
self._txs.add_edge(tx1, tx2, ddi=matches)
def test_printSwitches():
g = get_switches.GetSwitches(dataPath + 'annotation.pklz')
g.run(dataPath + 'switches')
io.printSwitches(g._genes, g._txs)
switches = [ x for x in io.readTable("switches_spada.tsv") ]
assert len(switches) == 8
os.remove("switches_spada.tsv")
def test_ppiAnalysis():
s = structural_analysis.StructuralAnalysis((g._genes, g._txs))
s.ppiAnalysis()
ddi = [x for x in io.readTable("ppi_analysis.tsv")]
assert len(ddi) == 6
assert len([x for x in ddi if x['What'] == "Unaffected"]) == 2
assert len([x for x in ddi if x['What'] == "Affected"]) == 2
assert len([x for x in ddi if x['What'] == "Lost_in_cases"]) == 1
assert len([x for x in ddi if x['What'] == "Gained_in_cases"]) == 1
os.remove("ppi_analysis.tsv")
def addAberrant(self, aberrant):
if aberrant:
self.logger.info("Import aberrant isoforms absent in GTF.")
prev = self._txs.skip_filter
self._txs.skip_filter = True
for line in io.readTable(aberrant, keys=['GeneId', 'Transcript']):
self._txs.add_node(line['Transcript'], line['GeneId'])
self._txs.update_node(line["Transcript"], "canonical", False)
self._txs.skip_filter = prev
def readSwitches(self, switchesFile, tx_network):
"""Import a set of genes with an isoform switch from candidateList.tsv.
"""
self.logger.debug("Retrieving calculated isoform switches.")
for line in io.readTable(switchesFile):
gene = line['GeneId']
ctrl = line['Control_transcript']
case = line['Case_transcript']
samples = set(line['Samples'].split(','))
if self.valid_switch(gene, ctrl, case, tx_network):
thisSwitch = LiteSwitch(ctrl, case, samples)
self.update_node("switches", thisSwitch, full_name=gene)
def getIsoformFeatures(self, features):
if self._new and not features:
raise SpadaError(
"A file with the protein features must be provided.")
elif not features:
self.logger.info(
"Protein features from the provided network will be used.")
return
self.logger.info("Reading isoform features.")
featureFields = [
'Transcript', 'Feature_type', 'Feature', 'Start', 'End'
]
for line in io.readTable(features, keys=featureFields):
tx = line['Transcript']
featureType = line['Feature_type']
feature = line['Feature']
start = int(line['Start'])
end = int(line['End'])
self._txs.update_node(tx, featureType, (start, end), feature)
def test_featureAnalysis():
s = structural_analysis.StructuralAnalysis((g._genes, g._txs))
s.featureAnalysis()
# pfams
pfam = [x for x in io.readTable("pfam_analysis.tsv")]
assert len(pfam) == 8
assert [
x for x in pfam
if x['Control_transcript'] == "ENST01.2" and x['Case_transcript'] ==
"ENST02.2" and x['What'] == "Nothing" and x['Feature'] == "D1"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST01.2" and x['Case_transcript'] ==
"ENST02.2" and x['What'] == "Lost_in_cases" and x['Feature'] == "D1"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST01.2" and x['Case_transcript'] ==
"ENST02.2" and x['What'] == "Gained_in_cases" and x['Feature'] == "D4"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST01.2" and x['Case_transcript'] ==
"ENST02.2" and x['What'] == "Gained_in_cases" and x['Feature'] == "D2"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST02.2" and x['Case_transcript'] ==
"ENST01.2" and x['What'] == "Nothing" and x['Feature'] == "D1"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST02.2" and x['Case_transcript'] ==
"ENST01.2" and x['What'] == "Gained_in_cases" and x['Feature'] == "D1"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST02.2" and x['Case_transcript'] ==
"ENST01.2" and x['What'] == "Lost_in_cases" and x['Feature'] == "D4"
]
assert [
x for x in pfam
if x['Control_transcript'] == "ENST02.2" and x['Case_transcript'] ==
"ENST01.2" and x['What'] == "Lost_in_cases" and x['Feature'] == "D2"
]
# prosite
prosite = [x for x in io.readTable("prosite_analysis.tsv")]
assert len(prosite) == 3
assert [
x for x in prosite
if x['Control_transcript'] == "ENST08.1" and x['Case_transcript'] ==
"ENST09.1" and x['What'] == "Nothing" and x['Feature'] == "P1"
]
assert [
x for x in prosite
if x['Control_transcript'] == "ENST08.1" and x['Case_transcript'] ==
"ENST09.1" and x['What'] == "Lost_in_cases" and x['Feature'] == "P1"
]
assert [
x for x in prosite
if x['Control_transcript'] == "ENST08.1" and x['Case_transcript'] ==
"ENST09.1" and x['What'] == "Gained_in_cases" and x['Feature'] == "P2"
]
# idr
idr = [x for x in io.readTable("idr_analysis.tsv")]
assert len(idr) == 6
assert [
x for x in idr
if x['Control_transcript'] == "ENST16.2" and x['Case_transcript'] ==
"ENST14.5" and x['What'] == "Lost_in_cases" and x['Sequence'] == "ABcd"
]
assert len([
x for x in idr if x['Control_transcript'] == "ENST16.2"
and x['Case_transcript'] == "ENST14.5"
]) == 1
os.remove("pfam_analysis.tsv")
os.remove("prosite_analysis.tsv")
os.remove("idr_analysis.tsv")