2 '''unit testing code for pysam.
4 Execute in the :file:`tests` directory as it requires the Makefile
5 and data files located there.
8 import sys, os, shutil, gzip
14 def checkBinaryEqual( filename1, filename2 ):
15 '''return true if the two files are binary equal.'''
16 if os.path.getsize( filename1 ) != os.path.getsize( filename2 ):
19 infile1 = open(filename1, "rb")
20 infile2 = open(filename2, "rb")
22 def chariter( infile ):
29 for c1,c2 in itertools.izip( chariter( infile1), chariter( infile2) ):
38 class TestIndexing(unittest.TestCase):
39 filename = "example.gtf.gz"
40 filename_idx = "example.gtf.gz.tbi"
44 self.tmpfilename = "tmp_%i.gtf.gz" % id(self)
45 shutil.copyfile( self.filename, self.tmpfilename )
47 def testIndexPreset( self ):
48 '''test indexing via preset.'''
50 pysam.tabix_index( self.tmpfilename, preset = "gff" )
51 checkBinaryEqual( self.tmpfilename + ".tbi", self.filename_idx )
54 os.unlink( self.tmpfilename )
55 os.unlink( self.tmpfilename + ".tbi" )
57 class TestCompression(unittest.TestCase):
58 filename = "example.gtf.gz"
59 filename_idx = "example.gtf.gz.tbi"
63 self.tmpfilename = "tmp_%i.gtf" % id(self)
64 infile = gzip.open( self.filename, "r")
65 outfile = open( self.tmpfilename, "w" )
66 outfile.write( "".join(infile.readlines()) )
70 def testIndexPreset( self ):
71 '''test indexing via preset.'''
73 pysam.tabix_index( self.tmpfilename, preset = "gff" )
74 checkBinaryEqual( self.tmpfilename + ".gz", self.filename )
75 checkBinaryEqual( self.tmpfilename + ".gz.tbi", self.filename_idx )
78 os.unlink( self.tmpfilename + ".gz" )
79 os.unlink( self.tmpfilename + ".gz.tbi" )
81 class TestIteration( unittest.TestCase ):
83 filename = "example.gtf.gz"
87 self.tabix = pysam.Tabixfile( self.filename )
88 lines = [ x for x in gzip.open(self.filename).readlines() if not x.startswith("#") ]
89 # creates index of contig, start, end, adds content without newline.
91 (x[0][0], int(x[0][3]), int(x[0][4]), x[1])
92 for x in [ (y.split("\t"), y[:-1]) for y in lines ] ]
94 def getSubset( self, contig = None, start = None, end = None):
98 subset = [ x[3] for x in self.compare ]
100 if start != None and end == None:
101 # until end of contig
102 subset = [ x[3] for x in self.compare if x[0] == contig and x[2] > start ]
103 elif start == None and end != None:
104 # from start of contig
105 subset = [ x[3] for x in self.compare if x[0] == contig and x[1] <= end ]
106 elif start == None and end == None:
107 subset = [ x[3] for x in self.compare if x[0] == contig ]
109 # all within interval
110 subset = [ x[3] for x in self.compare if x[0] == contig and \
111 min( x[2], end) - max(x[1], start) > 0 ]
115 def checkPairwise( self, result, ref ):
123 self.assertEqual( len(result), len(ref),
124 "unexpected number of results: %i, expected %i, differences are %s: %s" \
125 % (len(result), len(ref),
129 for x, d in enumerate( zip( result, ref )):
131 self.assertEqual( d[0], d[1],
132 "unexpected results in pair %i: '%s', expected '%s'" % \
139 result = list(self.tabix.fetch())
140 ref = self.getSubset( )
141 self.checkPairwise( result, ref )
143 def testPerContig( self ):
144 for contig in ("chr1", "chr2", "chr1", "chr2" ):
145 result = list(self.tabix.fetch( contig ))
146 ref = self.getSubset( contig )
147 self.checkPairwise( result, ref )
149 def testPerContigToEnd( self ):
152 for contig in ("chr1", "chr2", "chr1", "chr2" ):
153 for start in range( 0, 200000, 1000):
154 result = list(self.tabix.fetch( contig, start, end ))
155 ref = self.getSubset( contig, start, end )
156 self.checkPairwise( result, ref )
158 def testPerContigFromStart( self ):
161 for contig in ("chr1", "chr2", "chr1", "chr2" ):
162 for end in range( 0, 200000, 1000):
163 result = list(self.tabix.fetch( contig, start, end ))
164 ref = self.getSubset( contig, start, end )
165 self.checkPairwise( result, ref )
167 def testPerContig( self ):
169 start, end = None, None
170 for contig in ("chr1", "chr2", "chr1", "chr2" ):
171 result = list(self.tabix.fetch( contig, start, end ))
172 ref = self.getSubset( contig, start, end )
173 self.checkPairwise( result, ref )
175 def testPerInterval( self ):
177 start, end = None, None
178 for contig in ("chr1", "chr2", "chr1", "chr2" ):
179 for start in range( 0, 200000, 2000):
180 for end in range( start, start + 2000, 500):
181 result = list(self.tabix.fetch( contig, start, end ))
182 ref = self.getSubset( contig, start, end )
183 self.checkPairwise( result, ref )
186 def testInvalidIntervals( self ):
188 self.assertRaises( ValueError, self.tabix.fetch, "chr1", 0, -10)
189 self.assertRaises( ValueError, self.tabix.fetch, "chr1", -10, 200)
190 self.assertRaises( ValueError, self.tabix.fetch, "chr1", 200, 0)
191 self.assertRaises( ValueError, self.tabix.fetch, "chr1", -10, -20)
192 self.assertRaises( ValueError, self.tabix.fetch, "chrUn" )
194 def testGetContigs( self ):
195 self.assertEqual( sorted(self.tabix.contigs), ["chr1", "chr2"] )
196 # check that contigs is read-only
197 self.assertRaises( AttributeError, setattr, self.tabix, "contigs", ["chr1", "chr2"] )
199 def testHeader( self ):
201 for x in gzip.open( self.filename ):
202 if not x.startswith("#"): break
204 header = list( self.tabix.header )
205 self.assertEqual( ref, header )
207 class TestParser( unittest.TestCase ):
209 filename = "example.gtf.gz"
213 self.tabix = pysam.Tabixfile( self.filename )
214 self.compare = [ x[:-1].split("\t") for x in gzip.open( self.filename, "r") if not x.startswith("#") ]
216 def testRead( self ):
218 for x, r in enumerate(self.tabix.fetch( parser = pysam.asTuple() )):
219 self.assertEqual( self.compare[x], list(r) )
220 self.assertEqual( len(self.compare[x]), len(r) )
222 for c in range(0,len(r)):
223 self.assertEqual( self.compare[x][c], r[c] )
225 def testWrite( self ):
227 for x, r in enumerate(self.tabix.fetch( parser = pysam.asTuple() )):
228 self.assertEqual( self.compare[x], list(r) )
230 for y in range(len(r)):
231 r[y] = "test_%05i" % y
232 c[y] = "test_%05i" % y
233 self.assertEqual( c, list(r) )
234 self.assertEqual( "\t".join( c ), str(r) )
235 # check second assignment
236 for y in range(len(r)):
237 r[y] = "test_%05i" % y
238 self.assertEqual( c, list(r) )
239 self.assertEqual( "\t".join( c ), str(r) )
241 def testUnset( self ):
242 for x, r in enumerate(self.tabix.fetch( parser = pysam.asTuple() )):
243 self.assertEqual( self.compare[x], list(r) )
246 for y in range(len(r)):
249 self.assertEqual( c, list(r) )
250 self.assertEqual( "\t".join(e), str(r) )
252 class TestGTF( TestParser ):
253 def testRead( self ):
255 for x, r in enumerate(self.tabix.fetch( parser = pysam.asGTF() )):
256 self.assertEqual( "\t".join( self.compare[x]), str(r) )
258 class TestBed( unittest.TestCase ):
259 filename = "example.bed.gz"
263 self.tabix = pysam.Tabixfile( self.filename )
264 self.compare = [ x[:-1].split("\t") for x in gzip.open( self.filename, "r") if not x.startswith("#") ]
266 def testRead( self ):
268 for x, r in enumerate(self.tabix.fetch( parser = pysam.asBed() )):
270 self.assertEqual( "\t".join( c ), str(r) )
271 self.assertEqual( list(c), list(r) )
272 self.assertEqual( c[0], r.contig)
273 self.assertEqual( int(c[1]), r.start)
274 self.assertEqual( int(c[2]), r.end)
276 def testWrite( self ):
278 for x, r in enumerate(self.tabix.fetch( parser = pysam.asBed() )):
280 self.assertEqual( "\t".join( c ), str(r) )
281 self.assertEqual( list(c), list(r) )
284 self.assertEqual( "test", r.contig)
285 self.assertEqual( "test", r[0])
288 self.assertEqual( int(c[1]) + 1, r.start )
289 self.assertEqual( str(int(c[1]) + 1), r[1] )
292 self.assertEqual( int(c[2]) + 1, r.end )
293 self.assertEqual( str(int(c[2]) + 1), r[2] )
295 class TestVCF( TestParser ):
297 filename = "example.vcf40.gz"
298 columns = ("contig", "pos", "id",
299 "ref", "alt", "qual",
300 "filter", "info", "format" )
302 def testRead( self ):
304 ncolumns = len(self.columns)
306 for x, r in enumerate(self.tabix.fetch( parser = pysam.asVCF() )):
308 for y, field in enumerate( self.columns ):
310 self.assertEqual( int(c[y])-1, getattr( r, field ) )
311 self.assertEqual( int(c[y])-1, r.pos )
313 self.assertEqual( c[y], getattr( r, field ),
314 "mismatch in field %s: %s != %s" %\
315 ( field,c[y], getattr( r, field ) ) )
316 self.assertEqual( len(c), len( r ) + ncolumns )
318 for y in range(len(c) - ncolumns):
319 self.assertEqual( c[ncolumns+y], r[y] )
321 def testWrite( self ):
323 ncolumns = len(self.columns)
325 for x, r in enumerate(self.tabix.fetch( parser = pysam.asVCF() )):
327 for y, field in enumerate( self.columns ):
330 self.assertEqual( int(c[y]), getattr( r, field ) )
331 self.assertEqual( int(c[y]), r.pos )
333 setattr( r, field, "test_%i" % y)
335 self.assertEqual( c[y], getattr( r, field ),
336 "mismatch in field %s: %s != %s" %\
337 ( field,c[y], getattr( r, field ) ) )
339 self.assertEqual( len(c), len( r ) + ncolumns )
341 for y in range(len(c) - ncolumns):
342 c[ncolumns+y] = "test_%i" % y
344 self.assertEqual( c[ncolumns+y], r[y] )
346 if __name__ == "__main__":