# ---------------------------------------------------------------------------- # Copyright (c) 2019-2023, QIIME 2 development team. # # Distributed under the terms of the Modified BSD License. # # The full license is in the file LICENSE, distributed with this software. # ---------------------------------------------------------------------------- from qiime2.plugin.testing import TestPluginBase from qiime2.plugins import rescript import qiime2 import pandas as pd import numpy as np import pandas.util.testing as pdt import_data = qiime2.Artifact.import_data class TestMergeTaxa(TestPluginBase): package = 'rescript.tests' def setUp(self): super().setUp() self.merge_taxa = rescript.actions.merge_taxa m1 = pd.DataFrame({ 'Taxon': { # test merging full species vs. ambiguous level annotations '370253': 'k__Bacteria; p__Firmicutes; c__Clostridia; ' 'o__Clostridiales; f__Ruminococcaceae; ' 'g__Faecalibacterium; s__prausnitzii', # test merging identical annotations '2562097': 'k__Bacteria; p__Firmicutes; c__Bacilli; ' 'o__Bacillales; f__Bacillaceae; g__Bacillus; s__', # test merging incomplete vs. full annotations '2562091': 'k__Bacteria; p__Actinobacteria; c__Acidimicrobiia;' ' o__Acidimicrobiales; f__Microthrixaceae; g__; ' 's__', # test merging subset vs. superset of same lineage '4361279': 'k__Bacteria; p__Proteobacteria; ' 'c__Betaproteobacteria; o__Burkholderiales; ' 'f__Oxalobacteraceae; g__; s__', '4369464': 'k__Bacteria; p__; c__; o__; f__; g__; s__', # test merging unique feature (only found in dataframe a) 'unique2': 'k__Bacteria; p__; c__; o__; f__; g__; s__blah'}, 'confidence': { '370253': 1.0, '2562097': 0.99, '2562091': 1.0, '4361279': 0.9, '4369464': 0.8, 'unique2': 0.77}}) m1.index.name = 'Feature ID' self.m1 = import_data('FeatureData[Taxonomy]', pd.DataFrame(m1)) # Note: second column is "consensus" in frame b, but "confidence" in a m2 = pd.DataFrame({ 'Taxon': { '370253': 'k__Bacteria; p__Firmicutes; c__Clostridia; ' 'o__Clostridiales; f__; g__; s__', '2562097': 'k__Bacteria; p__Firmicutes; c__Bacilli; ' 'o__Bacillales; f__Bacillaceae; g__Bacillus; s__', '2562091': 'k__Bacteria; p__Actinobacteria; c__Acidimicrobiia;' ' o__Acidimicrobiales; f__', '4361279': 'k__Bacteria; p__Proteobacteria; ' 'c__Betaproteobacteria; o__Burkholderiales', '4369464': 'k__Bacteria; p__Proteobacteria; ' 'c__Alphaproteobacteria; o__Rhizobiales; ' 'f__Rhizobiaceae; g__Rhizobium; s__leguminosarum', # test merging unique feature (only found in dataframe b) 'unique1': 'k__Bacteria; p__; c__; o__; f__; g__; s__blah'}, 'Consensus': { '370253': 0.95, '2562097': 0.9, '2562091': 0.85, '4361279': 0.87, '4369464': 0.99, 'unique1': 0.75}}) m2.index.name = 'Feature ID' self.m2 = import_data('FeatureData[Taxonomy]', pd.DataFrame(m2)) # test merging entirely unique dataframe (no overlap with a or b) m3 = pd.DataFrame({ 'Taxon': { 'unique': 'k__Bacteria; p__; c__; o__; f__; g__; s__blah'}, 'confidence': { 'unique': 0.76}}) m3.index.name = 'Feature ID' self.m3 = import_data('FeatureData[Taxonomy]', pd.DataFrame(m3)) m4 = pd.DataFrame({ 'Taxon': { '370253': 'k__Protozoa; p__; c__; o__; f__; g__; s__blah', '2562097': 'k__Bacteria; p__Firmicutes; c__Bacilli; ' 'o__Bacillales; f__Bacillaceae; g__Bacillus; s__', '2562091': 'k__Bacteria; p__Actinobacteria; c__Acidimicrobiia;' ' o__Acidimicrobiales; f__', '4361279': 'k__Bacteria; p__Proteobacteria; ' 'c__Betaproteobacteria; o__Burkholderiales', '4369464': 'k__Bacteria; p__Proteobacteria; ' 'c__Alphaproteobacteria; o__Rhizobiales; ' 'f__Rhizobiaceae; g__Rhizobium; s__leguminosarum', # test merging unique feature (only found in dataframe b) 'unique1': 'k__Bacteria; p__; c__; o__; f__; g__; s__blah' }, 'consensus': { '370253': 0.3, '2562097': 0.9, '2562091': 0.85, '4361279': 0.87, '4369464': 0.99, 'unique1': 0.75 } }) m4.index.name = 'Feature ID' self.m4 = import_data('FeatureData[Taxonomy]', pd.DataFrame(m2)) # same as above, but these are single-column series self.s1 = import_data('FeatureData[Taxonomy]', m1['Taxon']) self.s2 = import_data('FeatureData[Taxonomy]', m2['Taxon']) self.s3 = import_data('FeatureData[Taxonomy]', m3['Taxon']) self.s4 = import_data('FeatureData[Taxonomy]', m4['Taxon']) def test_merge_taxa_lca(self): one_col, = self.merge_taxa([self.s1, self.s2, self.s3], 'lca', '', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales', '4369464': 'k__Bacteria', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( one_col.view(pd.DataFrame), exp, check_names=False) multi_col, = self.merge_taxa([self.m1, self.m2, self.m3], 'lca', '', ['disable']) pdt.assert_frame_equal( multi_col.view(pd.DataFrame), exp, check_names=False) def test_merge_taxa_lca_rank_handle(self): result, = self.merge_taxa( [self.s1, self.s2, self.s3], 'lca', '^[dkpcofgs]__', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'Bacteria;Actinobacteria;Acidimicrobiia;' 'Acidimicrobiales', '2562097': 'Bacteria;Firmicutes;Bacilli;Bacillales;' 'Bacillaceae;Bacillus;', '370253': 'Bacteria;Firmicutes;Clostridia;Clostridiales', '4361279': 'Bacteria;Proteobacteria;' 'Betaproteobacteria;Burkholderiales', '4369464': 'Bacteria', 'unique': 'Bacteria;;;;;;blah', 'unique1': 'Bacteria;;;;;;blah', 'unique2': 'Bacteria;;;;;;blah'}}) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False) def test_merge_taxa_lca_rank_handle_plus_new_rank_handle(self): result, = self.merge_taxa( [self.s1, self.s2, self.s3], 'lca', new_rank_handles=['kingdom', 'phylum', 'class', 'order', 'family', 'genus', 'species']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales', '4369464': 'k__Bacteria', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False) def test_merge_taxa_len(self): one_col, = self.merge_taxa([self.s1, self.s2, self.s3], 'len', '', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales;f__Microthrixaceae;g__;s__', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales;f__;g__;s__', '4361279': 'k__Bacteria;p__Proteobacteria;c__Betaproteobacteria;' 'o__Burkholderiales;f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria;p__Proteobacteria;c__Alphaproteobacteria;' 'o__Rhizobiales;f__Rhizobiaceae;g__Rhizobium;' 's__leguminosarum', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( one_col.view(pd.DataFrame), exp, check_names=False) exp = pd.concat([exp, pd.DataFrame({ 'Confidence': { '2562091': 1.0, '2562097': np.nan, '370253': np.nan, '4361279': 0.9, '4369464': np.nan, 'unique': 0.76, 'unique1': np.nan, 'unique2': 0.77}, 'Consensus': { '2562091': np.nan, '2562097': 0.9, '370253': 0.95, '4361279': np.nan, '4369464': 0.99, 'unique': np.nan, 'unique1': 0.75, 'unique2': np.nan} })], axis=1) multi_col, = self.merge_taxa([self.m1, self.m2, self.m3], 'len', '', ['disable']) multi_col = multi_col.view(pd.DataFrame).apply( lambda x: pd.to_numeric(x, errors='ignore')) pdt.assert_frame_equal(multi_col, exp, check_names=False) def test_merge_taxa_len_rank_handle(self): result, = self.merge_taxa( [self.m1, self.m2, self.m3], 'len', '^[dkpcofgs]__', ['disable']) exp = pd.DataFrame({ 'Taxon': { '2562091': 'Bacteria;Actinobacteria;Acidimicrobiia;' 'Acidimicrobiales;Microthrixaceae;;', '2562097': 'Bacteria;Firmicutes;Bacilli;Bacillales;' 'Bacillaceae;Bacillus;', # note how this expected taxon differs from that in mode=len # tests above; this is the benefit of rank_handle, that it # can pick the longest lineage after removing ambiguous labels '370253': 'Bacteria;Firmicutes;Clostridia;Clostridiales;' 'Ruminococcaceae;Faecalibacterium;prausnitzii', '4361279': 'Bacteria;Proteobacteria;Betaproteobacteria' ';Burkholderiales;Oxalobacteraceae;;', '4369464': 'Bacteria;Proteobacteria;Alphaproteobacteria;' 'Rhizobiales;Rhizobiaceae;Rhizobium;leguminosarum', 'unique': 'Bacteria;;;;;;blah', 'unique1': 'Bacteria;;;;;;blah', 'unique2': 'Bacteria;;;;;;blah'}, 'Confidence': {'2562091': 1.0, '2562097': np.nan, '370253': 1.0, '4361279': 0.9, '4369464': np.nan, 'unique': 0.76, 'unique1': np.nan, 'unique2': 0.77}, 'Consensus': {'2562091': np.nan, '2562097': 0.9, '370253': np.nan, '4361279': np.nan, '4369464': 0.99, 'unique': np.nan, 'unique1': 0.75, 'unique2': np.nan}}) result = result.view(pd.DataFrame).apply( lambda x: pd.to_numeric(x, errors='ignore')) pdt.assert_frame_equal(result, exp, check_names=False) def test_merge_taxa_len_rank_handle_plus_new_rank_handle(self): result, = self.merge_taxa( [self.m1, self.m2, self.m3], 'len', new_rank_handles=['kingdom', 'phylum', 'class', 'order', 'family', 'genus', 'species']) exp = pd.DataFrame({ 'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales;f__Microthrixaceae;g__;s__', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;' 'o__Bacillales;f__Bacillaceae;g__Bacillus;s__', # note how this expected taxon differs from that in mode=len # tests above; this is the benefit of rank_handle, that it # can pick the longest lineage after removing ambiguous labels '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales;f__Ruminococcaceae;' 'g__Faecalibacterium;s__prausnitzii', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales;' 'f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria;p__Proteobacteria;' 'c__Alphaproteobacteria;o__Rhizobiales;' 'f__Rhizobiaceae;g__Rhizobium;s__leguminosarum', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}, 'Confidence': {'2562091': 1.0, '2562097': np.nan, '370253': 1.0, '4361279': 0.9, '4369464': np.nan, 'unique': 0.76, 'unique1': np.nan, 'unique2': 0.77}, 'Consensus': {'2562091': np.nan, '2562097': 0.9, '370253': np.nan, '4361279': np.nan, '4369464': 0.99, 'unique': np.nan, 'unique1': 0.75, 'unique2': np.nan}}) result = result.view(pd.DataFrame).apply( lambda x: pd.to_numeric(x, errors='ignore')) pdt.assert_frame_equal(result, exp, check_names=False) # score should fail if there is only one column in any dataset def test_merge_taxa_score_one_column(self): with self.assertRaisesRegex(IndexError, "second column"): self.merge_taxa([self.m1, self.s2], 'score', '') def test_merge_taxa_score_multi_column(self): new, = self.merge_taxa([self.m1, self.m2, self.m3], 'score', '', ['disable']) exp = pd.DataFrame({ 'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales;f__Microthrixaceae;g__;s__', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales' ';f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales;f__Ruminococcaceae;' 'g__Faecalibacterium;s__prausnitzii', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales;' 'f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria;p__Proteobacteria;' 'c__Alphaproteobacteria;o__Rhizobiales;' 'f__Rhizobiaceae;g__Rhizobium;s__leguminosarum', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}, 'Confidence': { '2562091': 1.0, '2562097': 0.99, '370253': 1.0, '4361279': 0.9, '4369464': 0, 'unique': 0.76, 'unique1': 0, 'unique2': 0.77}, 'Consensus': { '2562091': 0, '2562097': 0, '370253': 0, '4361279': 0, '4369464': 0.99, 'unique': 0, 'unique1': 0.75, 'unique2': 0}, 'Score': { '2562091': 1.0, '2562097': 0.99, '370253': 1.0, '4361279': 0.9, '4369464': 0.99, 'unique': 0.76, 'unique1': 0.75, 'unique2': 0.77}}) new = new.view(pd.DataFrame).apply( lambda x: pd.to_numeric(x, errors='ignore')) pdt.assert_frame_equal(new, exp, check_names=False) # compare vs. LCA tests above to see super(set) mode function; the longer # taxonomy is selected if it is a superset of the other, otherwise LCA def test_merge_taxa_super_lca(self): one_col, = self.merge_taxa([self.s1, self.s2, self.s3], 'super', '', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales;f__Microthrixaceae;g__;s__', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales;f__Ruminococcaceae;' 'g__Faecalibacterium;s__prausnitzii', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales;' 'f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria;p__Proteobacteria;' 'c__Alphaproteobacteria;o__Rhizobiales;' 'f__Rhizobiaceae;g__Rhizobium;s__leguminosarum', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( one_col.view(pd.DataFrame), exp, check_names=False) multi_col, = self.merge_taxa([self.m1, self.m2, self.m3], 'super', '', ['disable']) pdt.assert_frame_equal( multi_col.view(pd.DataFrame), exp, check_names=False) # compare vs. LCA tests above to see super(set) mode function; the longer # taxonomy is selected if it is a superset of the other, otherwise LCA def test_merge_taxa_super_lca_rank_handle(self): result, = self.merge_taxa( [self.s1, self.s2, self.s3], 'super', '^[dkpcofgs]__', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'Bacteria;Actinobacteria;Acidimicrobiia;' 'Acidimicrobiales;Microthrixaceae;;', '2562097': 'Bacteria;Firmicutes;Bacilli;Bacillales;' 'Bacillaceae;Bacillus;', '370253': 'Bacteria;Firmicutes;Clostridia;Clostridiales;' 'Ruminococcaceae;Faecalibacterium;prausnitzii', '4361279': 'Bacteria;Proteobacteria;' 'Betaproteobacteria;Burkholderiales;Oxalobacteraceae;;', '4369464': 'Bacteria;Proteobacteria;Alphaproteobacteria;' 'Rhizobiales;Rhizobiaceae;Rhizobium;leguminosarum', 'unique': 'Bacteria;;;;;;blah', 'unique1': 'Bacteria;;;;;;blah', 'unique2': 'Bacteria;;;;;;blah'}}) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False) multi_col, = self.merge_taxa([self.m1, self.m2, self.m3], 'super', '^[dkpcofgs]__', ['disable']) pdt.assert_frame_equal( multi_col.view(pd.DataFrame), exp, check_names=False) def test_merge_taxa_super_lca_with_unassigned(self): result, = self.merge_taxa([self.s1, self.s4], 'super', '^[dkpcofgs]__', ['disable']) exp = pd.DataFrame({ 'Taxon': { '2562091': 'Bacteria;Actinobacteria;Acidimicrobiia;' 'Acidimicrobiales;Microthrixaceae;;', '2562097': 'Bacteria;Firmicutes;Bacilli;Bacillales;' 'Bacillaceae;Bacillus;', '370253': 'Unassigned', '4361279': 'Bacteria;Proteobacteria;' 'Betaproteobacteria;Burkholderiales;' 'Oxalobacteraceae;;', '4369464': 'Bacteria;Proteobacteria;Alphaproteobacteria;' 'Rhizobiales;Rhizobiaceae;Rhizobium;leguminosarum', 'unique1': 'Bacteria;;;;;;blah', 'unique2': 'Bacteria;;;;;;blah' } }) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False) # majority is super without superstring collapsing def test_merge_taxa_super_lca_majority(self): one_col, = self.merge_taxa([self.s1, self.s2, self.s3], 'majority', '', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales;' 'f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( one_col.view(pd.DataFrame), exp, check_names=False) multi_col, = self.merge_taxa( [self.m1, self.m2, self.m3], 'majority', '', ['disable']) pdt.assert_frame_equal( multi_col.view(pd.DataFrame), exp, check_names=False) # note that majority with rank handle behaves the same as super LCA above # (same expected outputs) since the super/substrings in the test data # consist of empty ranks vs. rank handles. def test_merge_taxa_majority_rank_handle(self): result, = self.merge_taxa( [self.s1, self.s2, self.s3], 'majority', '^[dkpcofgs]__', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'Bacteria;Actinobacteria;Acidimicrobiia;' 'Acidimicrobiales;Microthrixaceae;;', '2562097': 'Bacteria;Firmicutes;Bacilli;Bacillales;' 'Bacillaceae;Bacillus;', '370253': 'Bacteria;Firmicutes;Clostridia;Clostridiales;' 'Ruminococcaceae;Faecalibacterium;prausnitzii', '4361279': 'Bacteria;Proteobacteria;' 'Betaproteobacteria;Burkholderiales;Oxalobacteraceae;;', '4369464': 'Bacteria;Proteobacteria;Alphaproteobacteria;' 'Rhizobiales;Rhizobiaceae;Rhizobium;leguminosarum', 'unique': 'Bacteria;;;;;;blah', 'unique1': 'Bacteria;;;;;;blah', 'unique2': 'Bacteria;;;;;;blah'}}) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False) multi_col, = self.merge_taxa([self.m1, self.m2, self.m3], 'majority', '^[dkpcofgs]__', ['disable']) pdt.assert_frame_equal( multi_col.view(pd.DataFrame), exp, check_names=False) # make a more challenging test for majority with label redundancy and # no clear majority def test_merge_taxa_majority_rank_handle_true_majority(self): s4 = pd.DataFrame({ 'Taxon': { # ensure majority works (genus level) and at species level # LCA works on 3-way disagreement (no majority) '370253': 'k__Bacteria; p__Firmicutes; c__Clostridia; ' 'o__Clostridiales; f__Ruminococcaceae; ' 'g__Faecalibacterium; s__not_prausnitzii', # ensure that empty ranks are ignored by LCA majority '2562097': 'k__Bacteria; p__Firmicutes; c__Bacilli; ' 'o__Bacillales; f__Bacillaceae; g__Bacillus', # majority rules '2562091': 'k__Bacteria; p__Actinobacteria; c__Acidimicrobiia;' ' o__Acidimicrobiales; f__Microthrixaceae; g__; ' 's__'}}) s4.index.name = 'Feature ID' s4 = import_data('FeatureData[Taxonomy]', pd.DataFrame(s4)) result, = self.merge_taxa( [self.s1, self.s2, self.s3, s4], 'majority', '', ['disable']) exp = pd.DataFrame({'Taxon': { '2562091': 'k__Bacteria;p__Actinobacteria;c__Acidimicrobiia;' 'o__Acidimicrobiales;f__Microthrixaceae;g__;s__', '2562097': 'k__Bacteria;p__Firmicutes;c__Bacilli;o__Bacillales;' 'f__Bacillaceae;g__Bacillus;s__', '370253': 'k__Bacteria;p__Firmicutes;c__Clostridia;' 'o__Clostridiales;f__Ruminococcaceae;' 'g__Faecalibacterium', '4361279': 'k__Bacteria;p__Proteobacteria;' 'c__Betaproteobacteria;o__Burkholderiales;' 'f__Oxalobacteraceae;g__;s__', '4369464': 'k__Bacteria', 'unique': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique1': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah', 'unique2': 'k__Bacteria;p__;c__;o__;f__;g__;s__blah'}}) pdt.assert_frame_equal( result.view(pd.DataFrame), exp, check_names=False)