def verifyLineageSet(self, markerSetFile, bRequireTaxonomy):
"""Verify output of lineage set command."""
with open(markerSetFile) as f:
f.readline() # skip header
for line in f:
if line.strip() != '':
lineSplit = line.split('\t')
binId = lineSplit[0]
numMarkers = int(lineSplit[1])
uid = lineSplit[2]
lineage = lineSplit[3]
numGenomes = int(lineSplit[4])
markerSet = MarkerSet(uid, lineage, numGenomes, eval(lineSplit[5].rstrip()))
np.testing.assert_almost_equal(int(binId), 637000110, err_msg="Failed bin ID test")
if not bRequireTaxonomy:
# this might be a little unstable as it depends on HMMER and prodigal, but
# we will see how it goes
np.testing.assert_equal(markerSet.numSets(), 266, err_msg="Failed # marker set test")
np.testing.assert_equal(markerSet.numMarkers(), 2134, err_msg="Failed # markers test")
assert(uid == 'UID5199')
else:
np.testing.assert_equal(markerSet.numSets(), 282, err_msg="Failed # marker set test")
np.testing.assert_equal(markerSet.numMarkers(), 1254, err_msg="Failed # markers test")
assert(lineage == 'f__Enterobacteriaceae')
def testMarkerSet(self):
"""Verify marker set data structure."""
markers = [set(["a", "b"]), set(["c"])]
ms = MarkerSet(0, "k__Bacteria", 100, markers)
markerGenes, markerSets = ms.size()
self.assertEqual(markerGenes, 3)
self.assertEqual(markerSets, 2)
self.assertEqual(ms.numMarkers(), 3)
self.assertEqual(ms.numSets(), 2)
self.assertEqual(ms.getMarkerGenes(), set(["a", "b", "c"]))
def testMarkerSet(self):
"""Verify marker set data structure."""
markers = [set(['a', 'b']), set(['c'])]
ms = MarkerSet(0, 'k__Bacteria', 100, markers)
markerGenes, markerSets = ms.size()
self.assertEqual(markerGenes, 3)
self.assertEqual(markerSets, 2)
self.assertEqual(ms.numMarkers(), 3)
self.assertEqual(ms.numSets(), 2)
self.assertEqual(ms.getMarkerGenes(), set(['a', 'b', 'c']))
def testMarkerSet(self):
"""Verify marker set data structure."""
markers = [set(['a', 'b']), set(['c'])]
ms = MarkerSet(0, 'k__Bacteria', 100, markers)
markerGenes, markerSets = ms.size()
self.assertEqual(markerGenes, 3)
self.assertEqual(markerSets, 2)
self.assertEqual(ms.numMarkers(), 3)
self.assertEqual(ms.numSets(), 2)
self.assertEqual(ms.getMarkerGenes(), set(['a', 'b', 'c']))
def run(self):
# read internal nodes file
metadata = {}
for line in open('./experiments/classTree.internal_nodes.tsv'):
uid, label = [x.strip() for x in line.split('\t')]
metadata[uid] = label
# read all lineage-specific marker genes
treeParser = TreeParser()
uniqueIdToLineageStatistics = treeParser.readNodeMetadata()
for uid in metadata:
stats = uniqueIdToLineageStatistics[uid]
markerSet = MarkerSet(uid, 'NA', int(stats['# genomes']),
eval(stats['marker set']))
metadata[uid] += ' [%d, %d, %d]' % (stats['# genomes'],
markerSet.numMarkers(),
markerSet.numSets())
# write out results
fout = open('./experiments/classTree.internal_nodes.metadata.tsv', 'w')
for uid, label in metadata.items():
fout.write(uid + '\t' + label + '\n')
fout.close()
def run(self):
# read internal nodes file
metadata = {}
for line in open('./experiments/classTree.internal_nodes.tsv'):
uid, label = [x.strip() for x in line.split('\t')]
metadata[uid] = label
# read all lineage-specific marker genes
treeParser = TreeParser()
uniqueIdToLineageStatistics = treeParser.readNodeMetadata()
for uid in metadata:
stats = uniqueIdToLineageStatistics[uid]
markerSet = MarkerSet(uid, 'NA', int(stats['# genomes']), eval(stats['marker set']))
metadata[uid] += ' [%d, %d, %d]' % (stats['# genomes'], markerSet.numMarkers(), markerSet.numSets())
# write out results
fout = open('./experiments/classTree.internal_nodes.metadata.tsv', 'w')
for uid, label in metadata.iteritems():
fout.write(uid + '\t' + label + '\n')
fout.close()