def test_best_name_freqs_for_nodes(self):
"""correctly gets the frequencies per name per node"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a':['1','2','3','4','5','6','7'],
'b':['1','2','3','b','5','6','8'],
'd':['1','2','3','4','5','6','7'],
'e':['1','2','3','b','a','foo','7'],
'i':['1','2','3','4','a','foo','8'],
'j':['1','2','3','b','a','foo','8']}
total_counts = {0:{'1':10, 'foo':5},
1:{'2':10},
2:{'3':10},
3:{'4':4,'b':5},
4:{'5':7,'a':5},
5:{'6':3,'foo':3},
6:{'7':3,'8':4}}
tree = load_tree(input,tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
#result = best_name_freqs_for_nodes(tree)
cnode = tree.Children[0]
hnode = tree.Children[1]
knode = tree.Children[2]
def test_set_ranksafe(self):
"""correctly set ranksafe on tree"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a':['1','2','3','4','5','6','7'],
'b':['1','2','3','b','5','6','8'],
'd':['1','2','3','4','5','6','8'],
'e':['1','2','3','b','a','6','7'],
'i':['1','2','3','4','a','6','7'],
'j':['1','2','3','b','a','6','8']}
total_counts = {0:{'1':10, 'foo':5},
1:{'2':20},
2:{'3':10},
3:{'4':4,'b':5},
4:{'5':7,'a':5},
5:{'6':6},
6:{'7':3,'8':3}}
tree = load_tree(input, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
#exp_root = ['Yes','Majority','Yes','No','Yes','Yes','No']
exp_root = [True,False,True,False,True,True,False]
self.assertEqual(tree.RankSafe, exp_root)
def test_name_node_score_fold(self):
"""hate taxonomy"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a':['1','2','3','4','5','6','8'],
'b':['1','2','3','4','5','6','8'],
'd':['1','2','3','f','e','c','9'],
'e':['1','2','3','f','e','c','9'],
'i':['1','2','3','g','a','h','11'],
'j':['1','2','3','g','a','h','12']}
total_counts = {0:{'1':10, 'foo':5},
1:{'2':10},
2:{'3':10},
3:{'4':3,'f':5,'g':5},
4:{'5':7,'a':5,'e':4},
5:{'6':3,'c':3,'d':2,'h':3},
6:{'8':3,'9':2,'10':2,'11':2,'12':2}}
tree = load_tree(input,tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
name_node_score_fold(tree)
exp_root = ['1','2','3',None,None,None,None]
expc0 = [None, None, None, '4',None,None,None]
expc1 = [None, None, None, None,'e','c','9']
expc2 = [None, None, None, None, None,'h',None]
expc1c1 = [None] * 7
self.assertEqual(tree.RankNames, exp_root)
self.assertEqual(tree.Children[0].RankNames, expc0)
self.assertEqual(tree.Children[1].RankNames, expc1)
self.assertEqual(tree.Children[2].RankNames, expc2)
self.assertEqual(tree.Children[1].Children[1].RankNames, expc1c1)
def test_score_tree(self):
"""Determine's the tree's fmeasure score"""
# set RankNames and RankNameScores
# if name in RankNames, check score, look at tips, etc
t_str = "(((a,b),(c,d))e,(f,g),h)i;"
t = DndParser(t_str)
t.RankNames = ['i',None,None,None] # 1.0 * 6
t.RankNameScores = [1.0,None,None,None]
t.Children[0].RankNames = [None,'e','foo',None] # 0.5 * 3, 0.6 * 3
t.Children[0].RankNameScores = [None, 0.5, 0.6, None]
t.Children[0].Children[0].RankNames = [None] * 7
t.Children[0].Children[1].RankNames = [None] * 7
t.Children[1].RankNames = [None] * 7
t.Children[1].RankNameScores = [None] * 7
tips = t.tips()
tips[0].Consensus = [None] * 7
tips[1].Consensus = [1,3,None,None]
tips[2].Consensus = [2,4,5,None]
tips[3].Consensus = [None,1,None,None]
tips[4].Consensus = [None,1,None,None]
tips[5].Consensus = [2,None,3,None]
tips[6].Consensus = [None,4,None,None]
decorate_ntips(t)
exp = ((1.0 * 6) + (0.5 * 3) + (0.6 * 3)) / (6 + 3 + 3)
obs = score_tree(t)
self.assertEqual(obs, exp)
def test_decorate_name_relative_freqs(self):
"""correctly decorate relative frequency information on a tree"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a':['1','2','3','4','5','6','7'],
'b':['1','2','3','4','5','6','8'],
'd':['1','2','3','4','5','6','8'],
'e':['1','2','3','4','a','6','7'],
'i':['1','2','3','4','a',None,'7'],
'j':['1','2','3','4','a',None,'8']}
total_counts = {0:{'1':10, 'foo':5},
1:{'2':6},
2:{'3':12},
3:{'4':6,'bar':5},
4:{'5':6,'a':3},
5:{'6':6},
6:{'7':3,'8':3}}
tree = load_tree(input, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
exp_root = {0:{'1':.6},
1:{'2':1.0},
2:{'3':.5},
3:{'4':1.0},
4:{'5':.5, 'a':1.0},
5:{'6':4.0/6},
6:{'7':1.0,'8':1.0}}
self.assertFloatEqual(tree.ConsensusRelFreq, exp_root)
def test_score_tree(self):
"""Determine's the tree's fmeasure score"""
# set RankNames and RankNameScores
# if name in RankNames, check score, look at tips, etc
t_str = StringIO(u"(((a,b),(c,d))e,(f,g),h)i;")
t = TreeNode.read(t_str)
t.RankNames = ['i', None, None, None] # 1.0 * 6
t.RankNameScores = [1.0, None, None, None]
t.children[0].RankNames = [None, 'e', 'foo', None] # 0.5 * 3, 0.6 * 3
t.children[0].RankNameScores = [None, 0.5, 0.6, None]
t.children[0].children[0].RankNames = [None] * 7
t.children[0].children[1].RankNames = [None] * 7
t.children[1].RankNames = [None] * 7
t.children[1].RankNameScores = [None] * 7
tips = list(t.tips())
tips[0].Consensus = [None] * 7
tips[1].Consensus = [1, 3, None, None]
tips[2].Consensus = [2, 4, 5, None]
tips[3].Consensus = [None, 1, None, None]
tips[4].Consensus = [None, 1, None, None]
tips[5].Consensus = [2, None, 3, None]
tips[6].Consensus = [None, 4, None, None]
decorate_ntips(t)
exp = ((1.0 * 6) + (0.5 * 3) + (0.6 * 3)) / (6 + 3 + 3)
obs = score_tree(t)
self.assertEqual(obs, exp)
def test_decorate_name_relative_freqs(self):
"""correctly decorate relative frequency information on a tree"""
data = StringIO(u"((a,b)c,(d,(e,f)g)h,(i,j)k)l;")
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', '4', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '8'],
'e': ['1', '2', '3', '4', 'a', '6', '7'],
'i': ['1', '2', '3', '4', 'a', None, '7'],
'j': ['1', '2', '3', '4', 'a', None, '8']}
total_counts = {0: {'1': 10, 'foo': 5},
1: {'2': 6},
2: {'3': 12},
3: {'4': 6, 'bar': 5},
4: {'5': 6, 'a': 3},
5: {'6': 6},
6: {'7': 3, '8': 3}}
tree = load_tree(data, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
exp_root = {0: {'1': .6},
1: {'2': 1.0},
2: {'3': .5},
3: {'4': 1.0},
4: {'5': .5, 'a': 1.0},
5: {'6': 4.0 / 6},
6: {'7': 1.0, '8': 1.0}}
self.assertEqual(tree.ConsensusRelFreq, exp_root)
def test_pick_names(self):
"""correctly pick names to retain on a tree"""
data = StringIO(u"((a,b)c,(d,(e,f)g)h,(i,j)k)l;")
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', 'b', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '7'],
'e': ['1', '2', '3', 'b', 'a', 'foo', '7'],
'i': ['1', '2', '3', '4', 'a', 'foo', '8'],
'j': ['1', '2', '3', 'b', 'a', 'foo', '8']}
total_counts = {0: {'1': 10, 'foo': 5},
1: {'2': 10},
2: {'3': 10},
3: {'4': 4, 'b': 5},
4: {'5': 7, 'a': 5},
5: {'6': 3, 'foo': 3},
6: {'7': 3, '8': 4}}
tree = load_tree(data, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
exp_root = ['1', '2', '3', None, None, None, None]
self.assertEqual(tree.RankNames, exp_root)
expc0 = [None, None, None, None, None, '6', None]
expc1 = [None, None, None, None, None, None, '7']
expc2 = [None, None, None, None, None, 'foo', '8']
expc1c1 = [None] * 7
self.assertEqual(tree.children[0].RankNames, expc0)
self.assertEqual(tree.children[1].RankNames, expc1)
self.assertEqual(tree.children[2].RankNames, expc2)
self.assertEqual(tree.children[1].children[1].RankNames, expc1c1)
def test_best_name_freqs_for_nodes(self):
"""correctly gets the frequencies per name per node"""
data = StringIO(u"((a,b)c,(d,(e,f)g)h,(i,j)k)l;")
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', 'b', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '7'],
'e': ['1', '2', '3', 'b', 'a', 'foo', '7'],
'i': ['1', '2', '3', '4', 'a', 'foo', '8'],
'j': ['1', '2', '3', 'b', 'a', 'foo', '8']}
total_counts = {0: {'1': 10, 'foo': 5},
1: {'2': 10},
2: {'3': 10},
3: {'4': 4, 'b': 5},
4: {'5': 7, 'a': 5},
5: {'6': 3, 'foo': 3},
6: {'7': 3, '8': 4}}
tree = load_tree(data, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
# result = best_name_freqs_for_nodes(tree)
cnode = tree.children[0]
hnode = tree.children[1]
knode = tree.children[2]
def test_name_node_score_fold(self):
"""hate taxonomy"""
data = StringIO(u"((a,b)c,(d,(e,f)g)h,(i,j)k)l;")
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '8'],
'b': ['1', '2', '3', '4', '5', '6', '8'],
'd': ['1', '2', '3', 'f', 'e', 'c', '9'],
'e': ['1', '2', '3', 'f', 'e', 'c', '9'],
'i': ['1', '2', '3', 'g', 'a', 'h', '11'],
'j': ['1', '2', '3', 'g', 'a', 'h', '12']}
total_counts = {0: {'1': 10, 'foo': 5},
1: {'2': 10},
2: {'3': 10},
3: {'4': 3, 'f': 5, 'g': 5},
4: {'5': 7, 'a': 5, 'e': 4},
5: {'6': 3, 'c': 3, 'd': 2, 'h': 3},
6: {'8': 3, '9': 2, '10': 2, '11': 2, '12': 2}}
tree = load_tree(data, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
name_node_score_fold(tree)
exp_root = ['1', '2', '3', None, None, None, None]
expc0 = [None, None, None, '4', None, None, None]
expc1 = [None, None, None, None, 'e', 'c', '9']
expc2 = [None, None, None, None, None, 'h', None]
expc1c1 = [None] * 7
self.assertEqual(tree.RankNames, exp_root)
self.assertEqual(tree.children[0].RankNames, expc0)
self.assertEqual(tree.children[1].RankNames, expc1)
self.assertEqual(tree.children[2].RankNames, expc2)
self.assertEqual(tree.children[1].children[1].RankNames, expc1c1)
def test_set_ranksafe(self):
"""correctly set ranksafe on tree"""
data = StringIO(u"((a,b)c,(d,(e,f)g)h,(i,j)k)l;")
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', 'b', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '8'],
'e': ['1', '2', '3', 'b', 'a', '6', '7'],
'i': ['1', '2', '3', '4', 'a', '6', '7'],
'j': ['1', '2', '3', 'b', 'a', '6', '8']}
total_counts = {0: {'1': 10, 'foo': 5},
1: {'2': 20},
2: {'3': 10},
3: {'4': 4, 'b': 5},
4: {'5': 7, 'a': 5},
5: {'6': 6},
6: {'7': 3, '8': 3}}
tree = load_tree(data, tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
exp_root = [True, False, True, False, True, True, False]
self.assertEqual(tree.RankSafe, exp_root)
def test_pick_names(self):
"""correctly pick names to retain on a tree"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a':['1','2','3','4','5','6','7'],
'b':['1','2','3','b','5','6','8'],
'd':['1','2','3','4','5','6','7'],
'e':['1','2','3','b','a','foo','7'],
'i':['1','2','3','4','a','foo','8'],
'j':['1','2','3','b','a','foo','8']}
total_counts = {0:{'1':10, 'foo':5},
1:{'2':10},
2:{'3':10},
3:{'4':4,'b':5},
4:{'5':7,'a':5},
5:{'6':3,'foo':3},
6:{'7':3,'8':4}}
tree = load_tree(input,tipname_map)
decorate_ntips(tree)
decorate_name_relative_freqs(tree, total_counts, 1)
set_ranksafe(tree)
pick_names(tree)
exp_root = ['1','2','3',None,None,None,None]
self.assertEqual(tree.RankNames, exp_root)
expc0 = [None,None,None,None,None,'6',None]
expc1 = [None,None,None,None,None,None,'7']
expc2 = [None,None,None,None,None,'foo','8']
expc1c1 = [None] * 7
self.assertEqual(tree.Children[0].RankNames, expc0)
self.assertEqual(tree.Children[1].RankNames, expc1)
self.assertEqual(tree.Children[2].RankNames, expc2)
self.assertEqual(tree.Children[1].Children[1].RankNames, expc1c1)
def test_decorate_ntips(self):
"""correctly decorate the tree with the NumTips param"""
data = StringIO(u"(((a,b)c,(d,e,f)g)h,(i,j)k)l;")
tree = TreeNode.read(data)
tips = dict([(tip.name, tip) for tip in tree.tips()])
tips['a'].Consensus = [1, 2, 3, 4, 5, 6, 7]
tips['b'].Consensus = [None, None, None, 5, None, None, None]
tips['d'].Consensus = [1, 2, 3, 4, 5, 6, 8]
tips['e'].Consensus = [None, None, None, None, None, None, None]
tips['f'].Consensus = [1, 2, 3, 4, 5, 6, 8]
tips['i'].Consensus = [1, 2, 3, 4, 5, 6, 8]
tips['j'].Consensus = [1, 2, 3, 4, 5, 6, 8]
decorate_ntips(tree)
self.assertEqual(tree.NumTips, 6)
self.assertEqual(tree.children[0].NumTips, 4)
self.assertEqual(tree.children[1].NumTips, 2)
self.assertEqual(tree.children[0].children[0].NumTips, 2)
self.assertEqual(tree.children[0].children[1].NumTips, 2)
def test_decorate_ntips(self):
"""correctly decorate the tree with the NumTips param"""
input = "(((a,b)c,(d,e,f)g)h,(i,j)k)l;"
tree = DndParser(input)
tips = dict([(tip.Name, tip) for tip in tree.tips()])
tips['a'].Consensus = [1,2,3,4,5,6,7]
tips['b'].Consensus = [None,None,None,5,None,None,None]
tips['d'].Consensus = [1,2,3,4,5,6,8]
tips['e'].Consensus = [None, None,None,None,None,None,None]
tips['f'].Consensus = [1,2,3,4,5,6,8]
tips['i'].Consensus = [1,2,3,4,5,6,8]
tips['j'].Consensus = [1,2,3,4,5,6,8]
decorate_ntips(tree)
self.assertEqual(tree.NumTips, 6)
self.assertEqual(tree.Children[0].NumTips, 4)
self.assertEqual(tree.Children[1].NumTips, 2)
self.assertEqual(tree.Children[0].Children[0].NumTips, 2)
self.assertEqual(tree.Children[0].Children[1].NumTips, 2)
def test_decorate_name_counts(self):
"""correctly decorate relative frequency information on a tree"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {
'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', '4', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '8'],
'e': ['1', '2', '3', '4', 'a', '6', '7'],
'i': ['1', '2', '3', '4', 'a', None, '7'],
'j': ['1', '2', '3', '4', 'a', None, '8']
}
tree = load_tree(input, tipname_map)
decorate_ntips(tree)
decorate_name_counts(tree)
exp_root = {
0: {
'1': 6
},
1: {
'2': 6
},
2: {
'3': 6
},
3: {
'4': 6
},
4: {
'5': 3,
'a': 3
},
5: {
'6': 4
},
6: {
'7': 3,
'8': 3
}
}
self.assertEqual(tree.TaxaCount, exp_root)
def test_decorate_name_counts(self):
"""correctly decorate relative frequency information on a tree"""
input = "((a,b)c,(d,(e,f)g)h,(i,j)k)l;"
tipname_map = {'a': ['1', '2', '3', '4', '5', '6', '7'],
'b': ['1', '2', '3', '4', '5', '6', '8'],
'd': ['1', '2', '3', '4', '5', '6', '8'],
'e': ['1', '2', '3', '4', 'a', '6', '7'],
'i': ['1', '2', '3', '4', 'a', None, '7'],
'j': ['1', '2', '3', '4', 'a', None, '8']}
tree = load_tree(input, tipname_map)
decorate_ntips(tree)
decorate_name_counts(tree)
exp_root = {0: {'1': 6},
1: {'2': 6},
2: {'3': 6},
3: {'4': 6},
4: {'5': 3, 'a': 3},
5: {'6': 4},
6: {'7': 3, '8': 3}}
self.assertEqual(tree.TaxaCount, exp_root)
def generate_constrings(tree, tipname_map, verbose=False):
"""Assigns taxonomy to unidentified sequences in tree.
Returns all sequence IDs on tree."""
counts = nlevel.collect_names_at_ranks_counts(tree)
min_count = 2
nlevel.decorate_ntips(tree)
nlevel.decorate_name_relative_freqs(tree, counts, min_count)
nlevel.set_ranksafe(tree)
nlevel.pick_names(tree)
nlevel.name_node_score_fold(tree)
if verbose:
print "Tree score: ", nlevel.score_tree(tree)
nlevel.set_preliminary_name_and_rank(tree)
contree, contree_lookup = nlevel.make_consensus_tree(tipname_map.values())
nlevel.backfill_names_gap(tree, contree_lookup)
nlevel.commonname_promotion(tree)
nlevel.make_names_unique(tree, append_suffix=False)
constrings = nlevel.pull_consensus_strings(tree)
return constrings