def test_filter_on_strand(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
rna_plus = rna_seq.filterByStrand('+')
self.assertEqual(len(rna_plus),5,
"Wrong number of + strands")
rna_minus = rna_seq.filterByStrand('-')
self.assertEqual(len(rna_minus),5,
"Wrong number of - strands")
def test_filter_on_chromosome(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
rna_chr = rna_seq.filterByChr('chr3LHet')
self.assertEqual(len(rna_chr),1,
"Wrong number of chromosomes")
for rna_data in rna_chr:
self.assertEqual(rna_data.chrom,'chr3LHet',
"Wrong chromosome filtered")
def test_filter_on_TSS(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
lower,upper = 5000000,10000000
rna_tss = rna_seq.filterByTSS(upper,lower)
self.assertEqual(len(rna_tss),3,
"Wrong number of transcripts filtered on TSS")
for rna_data in rna_tss:
self.assertTrue((rna_data.getTSS() >= lower and
rna_data.getTSS() <= upper),
"Transcript outside range")
def test_getTSS(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
rna_plus = rna_seq.filterByStrand('+')
for rna_data in rna_plus:
self.assertTrue((rna_data.strand == '+' and
rna_data.start == rna_data.getTSS()),
"Incorrect TSS on + strand")
rna_minus = rna_seq.filterByStrand('-')
for rna_data in rna_minus:
self.assertTrue((rna_data.strand == '-' and
rna_data.end == rna_data.getTSS()),
"Incorrect TSS on - strand")
def test_write_features_summary(self):
# Set up some test data
peak = Peak('chr2L',66811,66812)
features = FeatureSet(
features_list=(
Feature('CG31973','chr2L',25402,59243,'-'),
Feature('CG2674-RE','chr2L',106903,114433,'+'),
Feature('CG2674-RC','chr2L',107926,114433,'+')))
# Temp output file
fp,summary = tempfile.mkstemp()
# Write peaks to file
ap = AnalysisReportWriter(output.MULTI_LINE,
fields=('peak.chr',
'peak.start',
'peak.end',
'order',
'feature.id',
'dist_closest',
'dist_TSS'),
summary=summary)
ap.write_nearest_features(peak,features)
ap.close()
# Expected and actual output
expected_output = \
"#peak.chr\tpeak.start\tpeak.end\torder\tfeature.id\tdist_closest\tdist_TSS\n" \
"chr2L\t66811\t66812\t1 of 3\tCG31973\t7568\t7568\n"
actual_output = open(summary,'r').read()
# Check that output matches
self.assertEqual(expected_output,actual_output)
def setUp(self):
# Set up some test data
self.peak = Peak('chr2L',66811,66812)
self.features = FeatureSet(
features_list=(
Feature('CG31973','chr2L',25402,59243,'-'),
Feature('CG2674-RE','chr2L',106903,114433,'+'),
Feature('CG2674-RC','chr2L',107926,114433,'+')))
self.single_line_fields = ('peak.chr',
'peak.start',
'peak.end',
'number_of_results',
'list(feature.id,'
'dist_closest,dist_TSS)')
self.single_line_fields_extra_data = ('peak.chr',
'peak.start',
'peak.end',
'cutoff',
'number_of_results',
'list(feature.id,'
'dist_closest,dist_TSS)')
self.multi_line_fields = ('peak.chr',
'peak.start',
'peak.end',
'order',
'feature.id',
'dist_closest',
'dist_TSS')
def test_write_features_append(self):
# Set up some test data
peak = Peak('chr2L',66811,66812)
features1 = FeatureSet(
features_list=(
Feature('CG31973','chr2L',25402,59243,'-'),
Feature('CG2674-RE','chr2L',106903,114433,'+'),))
features2 = FeatureSet(
features_list=(
Feature('CG2674-RC','chr2L',107926,114433,'+'),))
# Temp output file
fp,outfile = tempfile.mkstemp()
# Write first set of nearest features
ap = AnalysisReportWriter(output.MULTI_LINE,
fields=('peak.chr',
'peak.start',
'peak.end',
'order',
'feature.id',
'dist_closest',
'dist_TSS'),
outfile=outfile)
ap.write_nearest_features(peak,features1)
ap.close()
# Write second set of nearest features
ap = AnalysisReportWriter(output.MULTI_LINE,
fields=('peak.chr',
'peak.start',
'peak.end',
'order',
'feature.id',
'dist_closest',
'dist_TSS'),
outfile=outfile,
append=True)
ap.write_nearest_features(peak,features2)
ap.close()
# Expected and actual output
expected_output = \
"#peak.chr\tpeak.start\tpeak.end\torder\tfeature.id\tdist_closest\tdist_TSS\n" \
"chr2L\t66811\t66812\t1 of 2\tCG31973\t7568\t7568\n" \
"chr2L\t66811\t66812\t2 of 2\tCG2674-RE\t40091\t40091\n" \
"chr2L\t66811\t66812\t1 of 1\tCG2674-RC\t41114\t41114\n"
actual_output = open(outfile,'r').read()
# Check that output matches
self.assertEqual(expected_output,actual_output)
def test_sort_by_distance(self):
rna_sort = FeatureSet('Transcripts-ex1.txt')
position = 4250000
# Do sort on distance
# Sort is done in place, so assignment is not required
# however the sort function should return a reference to
# the initial object
result = rna_sort.sortByDistanceFrom(position)
self.assertEqual(result,rna_sort,
"Returned object doesn't match subject")
# Check that each distance is greater than the previous one
last_rna_data = None
for rna_data in rna_sort:
if not last_rna_data:
last_rna_data = rna_data
else:
self.assertTrue((abs(rna_data.getTSS() - position) >=
abs(last_rna_data.getTSS() - position)),
"Sort by distance failed")
def test_sort_by_closest_TSS_to_edge(self):
rna_sort = FeatureSet('Transcripts-ex1.txt')
position = (16000000,17500000)
# Do sort
# Sort is done in place, so assignment is not required
# however the sort function should return a reference to
# the initial object
result = rna_sort.sortByClosestTSSTo(*position)
self.assertEqual(result,rna_sort,
"Returned object doesn't match subject")
# Check that the closest distances are in ascending order
last_rna_data = None
for rna_data in rna_sort:
if not last_rna_data:
last_rna_data = rna_data
else:
self.assertTrue((min(abs(rna_data.getTSS() - position[0]),
abs(rna_data.getTSS() - position[1])) >=
min(abs(last_rna_data.getTSS() - position[0]),
abs(last_rna_data.getTSS() - position[1]))),
"Sort by closest TSS to edge failed")
def test_closest_transcript_to_peak(self):
features = FeatureSet('Transcripts-ex1.txt')
feature1 = features[1]
feature2 = features[2]
peaks = PeakSet('ChIP_peaks-ex1.txt')
peak = peaks[0]
nearest = GetNearestTranscriptToPeak(feature1, feature2, peak)
# 2nd feature should be closer than first
self.assertEqual(nearest, feature1,
"Wrong transcript selected as nearest")
# test when only one is set
nearest = GetNearestTranscriptToPeak(None, feature2, peak)
self.assertEqual(nearest, feature2,
"Wrong transcript selected as nearest")
def test__eq__(self):
# Check equality of FeatureSets
feature_set1 = FeatureSet()
feature_set2 = FeatureSet()
# Empty feature sets
self.assertEqual(feature_set1,feature_set2)
# Populate
feature_set1.addFeature(Feature('CG1000','chr1','1','2','+'))
feature_set2.addFeature(Feature('CG1000','chr1','1','2','+'))
self.assertEqual(feature_set1,feature_set2)
# Add second
feature_set1.addFeature(Feature('CG2000','chr1','1','2','+'))
self.assertNotEqual(feature_set1,feature_set2)
feature_set2.addFeature(Feature('CG2000','chr1','1','2','+'))
self.assertEqual(feature_set1,feature_set2)
# Add third
feature_set1.addFeature(Feature('CG2001','chr2',3,4,'-'))
feature_set2.addFeature(Feature('CG2002','chr2',3,5,'+'))
self.assertNotEqual(feature_set1,feature_set2)
def test_filter_on_flag(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
rna_flagged = rna_seq.filterByFlag(1)
self.assertEqual(len(rna_flagged),4,
"Wrong number of flagged data lines")
def test_reading_in_RNAseq_data(self):
rna_seq = FeatureSet('Transcripts-ex1.txt')
self.assertEqual(len(rna_seq),10,
"Wrong number of lines from RNA-seq file")
self.assertTrue(rna_seq.isFlagged(),
"Data should be flagged")
