Python TreeNode.read方法代码示例

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def depth_partition(self, input_tree, percentile, output_tree):
     '''
     Attempt to cluster tree with nodes of tip-to-tip distrubution <
     an nth percentile cutoff of the whole-tree distance distribution. 
     A better description can be found in the citation below.
     
     Parameters
     ----------
     tree: skbio TreeNode obj
         http://scikit-bio.org/docs/latest/generated/skbio.tree.TreeNode.html #skbio.tree.TreeNode
             
     percentile: float
         The percentile cutoff to use to determine the cutoff from clading
         from a given node.
     
     Clustering method modified from Prosperi et al method:
     Prosperi, M.C.F., et al. A novel methodology for large-scale phylogeny 
     partition. Nat. Commun. 2:321 doi: 10.1038/ncomms1325 (2011).
     
     http://www.nature.com/ncomms/journal/v2/n5/full/ncomms1325.html
     '''
     tree = TreeNode.read(input_tree)
     tree = tree.root_at_midpoint()
     cluster_count = 1
     clustered = set()
     clusters = {}
     logging.debug("Calculating %ith percentile cutoff from root" \
                                             % (percentile))
     whole_tree_distribution = self._node_dist(tree)
     
     cutoff = np.percentile(whole_tree_distribution, percentile)
     logging.debug("Cutoff (%ith percentile): %f" % (percentile,
                                                     cutoff))
     for node in tree.preorder():
         if node in clustered:
             continue
         elif node.is_tip():
             continue
         else:
             node_distribution = self._node_dist(node)
             median=np.median(node_distribution)
             logging.debug("Median of node: %f" % median)
             if median <= cutoff:
                 logging.debug("Cluster found!")
                 cluster_name =  "partition_%i" % (cluster_count)
                 clusters[cluster_name] = [x.name.replace(' ','_') 
                                           for x in node.tips()]
                 self._rename(node, cluster_name)
                 cluster_count+=1
                 for descenent in node.traverse():
                     clustered.add(descenent)
     logging.info("%i depth cluster(s) found in tree" % (cluster_count-1))
     tree.write(output_tree, "newick")
     
     logging.debug("Recording tips that were not partitioned")
     clusters[self.UNCLUSTERED] = []
     for tip in tree.tips():
         if tip not in clustered:
             clusters[self.UNCLUSTERED].append(tip.name.replace(' ','_'))
     return clusters

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testFindParents(self):
     ann = TreeAnnotator()
     
     tree = TreeNode.read(StringIO("(((A:1, B:2)'g__genus1':3, (C:4, D:5)'g__genus2':6)'f__family':10)root;"))
     assert_equals('g__genus1', ann.find_named_parent(tree, tree.find('B')).name, 'self is named')
     
     tree = TreeNode.read(StringIO("(((A:1, 2475:2)'g__genus1':3, (C:4, D:5)'g__genus2':6)'f__family':10)root;"))
     assert_equals('g__genus1', ann.find_named_parent(tree, tree.find('2475')).name, 'parent directly above')
     
     tree = TreeNode.read(StringIO("(((A:1, 2475:2):3, (C:4, D:5)'g__genus2':6)'f__family':10)root;"))
     assert_equals('f__family', ann.find_named_parent(tree, tree.find('2475')).name, 'parent 2 above')
     
     tree = TreeNode.read(StringIO("(((A:1, 2475:2):3, (C:4, D:5)'g__genus2':6)'f__family':10);"))
     assert_equals(None, ann.find_named_parent(tree, tree.find('f__family').parent), 'parent of root')
     
     tree = TreeNode.read(StringIO("(((A:1, 2475:2):3, (C:4, D:5)'g__genus2':6):10);"))
     assert_equals(None, ann.find_named_parent(tree, tree.find('g__genus2').parent), 'no parent before root')

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
def generate_html_summary(qclient, job_id, parameters, out_dir):
    """Generates the HTML summary of a BIOM artifact

    Parameters
    ----------
    qclient : qiita_client.QiitaClient
        The Qiita server client
    job_id : str
        The job id
    parameters : dict
        The parameter values to validate and create the artifact
    out_dir : str
        The path to the job's output directory

    Returns
    -------
    bool, None, str
        Whether the job is successful
        Ignored
        The error message, if not successful
    """
    # Step 1: gather file information from qiita using REST api
    artifact_id = parameters['input_data']
    qclient_url = "/qiita_db/artifacts/%s/" % artifact_id
    artifact_info = qclient.get(qclient_url)

    # Step 2: get the mapping file, depends if analysis or not
    if artifact_info['analysis'] is None:
        is_analysis = False
        qurl = ('/qiita_db/prep_template/%s/' %
                artifact_info['prep_information'][0])
        md = qclient.get(qurl)['qiime-map']
    else:
        is_analysis = True
        qurl = '/qiita_db/analysis/%s/metadata/' % artifact_info['analysis']
        md = qclient.get(qurl)

    tree = None
    if 'plain_text' in artifact_info['files']:
        tree = TreeNode.read(artifact_info['files']['plain_text'][0])

    # Step 3: generate HTML summary
    # if we get to this point of the code we are sure that this is a biom file
    # and that it only has one element
    index_fp, viz_fp, qza_fp = _generate_html_summary(
        artifact_info['files']['biom'][0], md, out_dir, is_analysis, tree)

    # Step 4: add the new file to the artifact using REST api
    success = True
    error_msg = ""
    try:
        qclient.patch(qclient_url, 'add', '/html_summary/',
                      value=dumps({'html': index_fp, 'dir': viz_fp}))
    except Exception as e:
        success = False
        error_msg = str(e)

    return success, None, error_msg

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
    def testTipToCluster(self):
        tree = TreeNode.read(StringIO('((F:20, ((A:11, B:12):10, (H:8, D:9):3):20)G:30)root;'))
        clusters = Tree2Tax().named_clusters_for_several_thresholds(tree, [40, 25])
        self.assertSameClusterSets([[25,[['F'], _('A B'), _('D H')]], [40,[['F'], _('A B D H')]]], clusters)
        assert_equals(_('G.3 G.1 G.2'), [c.name() for c in clusters[0].clusters])
        assert_equals(_('G.2 G.1'), [c.name() for c in clusters[1].clusters])
         
        tip = tree.find('F')
        assert_equals('G.3', clusters[0].tip_to_cluster(tip).name())
        assert_equals('G.2', clusters[1].tip_to_cluster(tip).name())
 
        tip = tree.find('D')
        assert_equals('G.2', clusters[0].tip_to_cluster(tip).name())
        assert_equals('G.1', clusters[1].tip_to_cluster(tip).name())

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
    def _open_tree(self, tree_path):
        '''
        Open a tree file, determine what decorations are already present. Strip
        Unwanted decoration
        
        Parameters
        ----------
        tree_path: str
            Path to a file containing a phylogenetic tree, in Newick format.
            
        Returns
        -------
        skbio TreeNode object
        '''
        tree_obj=TreeNode.read(open(tree_path))
        bootstrapped = True
        for node in tree_obj.non_tips():
            if node.name:
                try:
                    float(node.name)
                except:
                    logging.debug("Tree is decorated already. Stripping all \
    previous decoration from the tree.")
                    bootstrapped = False
                    tree_obj = self._strip_tree(tree_obj)
                    break
            else:
                if bootstrapped:
                    logging.warning("This tree doesn't appear correctly \
formatted or there is information missing. No boostrap value or decoration \
found for bare node. ")
                    bootstrapped = False
        if bootstrapped:
            logging.debug("Tree is bootstrap or has confidence values \
assigned to the nodes.")
        return tree_obj

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testClusterNamingWithBootstraps(self):
     tree = TreeNode.read(StringIO("((F:20, ((A:11, B:12):10, (H:8, D:9):3):20)'0.7:G':30)root;"))
     clusters = Tree2Tax().named_clusters(tree, 40)
     self.assertSameClusters([['F'], _('A B D H')], clusters)
     assert_equals(_('G.2 G.1'), [c.name() for c in clusters])

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testNamingWithBootstraps(self):
     tree = TreeNode.read(StringIO('((A:0.11, B:0.12)0.091:0.1, D:0.2)root;'))
     clusters = Tree2Tax().named_clusters(tree, 0.05)
     self.assertSameClusters([['A'],['B'],['D']], clusters)
     assert_equals(_('Root.1 Root.2 Root.3'), [c.name() for c in clusters])

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testClusterNamingOnTwoInternalNodesReverseOrder(self):
     tree = TreeNode.read(StringIO('((F:20, ((A:11, B:12):10, (H:8, D:9):3):20)G:30)root;'))
     clusters = Tree2Tax().named_clusters(tree, 40)
     self.assertSameClusters([['F'], _('A B D H')], clusters)
     assert_equals(_('G.2 G.1'), [c.name() for c in clusters])

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testClusterIntoThree(self):
     tree = TreeNode.read(StringIO('((((A:11, B:12)C:10, (H:8, D:9)I:3)E:20, F:20)G:30)root;'))
     clusters = Tree2Tax().named_clusters(tree, 25)
     self.assertSameClusters([_('A B'), _('D H'), ['F']], clusters)

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testClusterOnTwoInternalNodes(self):
     tree = TreeNode.read(StringIO('((((A:11, B:12)C:10, (H:8, D:9)I:3)E:20, F:20)G:30)root;'))
     clusters = Tree2Tax().named_clusters(tree, 40)
     self.assertSameClusters([_('A B D H'), ['F']], clusters)

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testTreeSubtree2(self):
     '''one genus is a subtree of another, and the longest branch is in both subtrees'''
     tree = TreeNode.read(StringIO("((((A:1, B:52)'g__genus1':3, D:50)'g__genus2':6)'f__family':10)root;"))
     examples = ThresholdFinder().find_examples(tree, 'f', 'g')
     self.assertSameCladeDistanceSet([['f__family','g__genus1','g__genus2',105.0]],
                                      examples)

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]

#.........这里部分代码省略.........
    table = load_table(biom_fp)
    metadata_ids = set(metadata)
    biom_sample_ids = set(table.ids())

    if not metadata_ids.issuperset(biom_sample_ids):
        # The BIOM sample ids are different from the ones in the prep template
        qclient.update_job_step(job_id, "Step 3: Fixing BIOM sample ids")
        # Attempt 1: the user provided the run prefix column - in this case
        # the run prefix column holds the sample ids present in the BIOM file
        if 'run_prefix' in metadata[next(iter(metadata_ids))]:
            id_map = {v['run_prefix']: k for k, v in metadata.items()}
        else:
            # Attemp 2: the sample ids in the BIOM table are the same that in
            # the prep template but without the prefix
            prefix = next(iter(metadata_ids)).split('.', 1)[0]
            prefixed = set("%s.%s" % (prefix, s) for s in biom_sample_ids)
            if metadata_ids.issuperset(prefixed):
                id_map = {s: "%s.%s" % (prefix, s) for s in biom_sample_ids}
            else:
                # There is nothing we can do. The samples in the BIOM table do
                # not match the ones in the prep template and we can't fix it
                error_msg = ('The sample ids in the BIOM table do not match '
                             'the ones in the prep information. Please, '
                             'provide the column "run_prefix" in the prep '
                             'information to map the existing sample ids to '
                             'the prep information sample ids.')
                return False, None, error_msg

        # Fix the sample ids
        try:
            table.update_ids(id_map, axis='sample')
        except TableException:
            missing = biom_sample_ids - set(id_map)
            error_msg = ('Your prep information is missing samples that are '
                         'present in your BIOM table: %s' % ', '.join(missing))
            return False, None, error_msg

        new_biom_fp = join(out_dir, basename(biom_fp))
        with biom_open(new_biom_fp, 'w') as f:
            table.to_hdf5(f, "Qiita BIOM type plugin")

    filepaths = [(new_biom_fp, 'biom')]

    # Validate the representative set, if it exists
    if 'preprocessed_fasta' in files:
        repset_fp = files['preprocessed_fasta'][0]

        # The observations ids of the biom table should be the same
        # as the representative sequences ids found in the representative set
        observation_ids = table.ids(axis='observation').tolist()
        extra_ids = []
        for record in load([repset_fp], constructor=FastaIterator):
            rec_id = record['SequenceID'].split()[0]
            try:
                observation_ids.remove(rec_id)
            except ValueError:
                extra_ids.append(rec_id)

        error_msg = []
        if extra_ids:
            error_msg.append("The representative set sequence file includes "
                             "observations not found in the BIOM table: %s"
                             % ', '.join(extra_ids))
        if observation_ids:
            error_msg.append("The representative set sequence file is missing "
                             "observation ids found in the BIOM tabe: %s" %
                             ', '.join(observation_ids))

        if error_msg:
            return False, None, '\n'.join(error_msg)

        filepaths.append((repset_fp, 'preprocessed_fasta'))

    # Validate the sequence specific phylogenetic tree (e.g. generated
    # by SEPP for Deblur), if it exists
    tree = None
    if 'plain_text' in files:
        phylogeny_fp = files['plain_text'][0]

        try:
            tree = TreeNode.read(phylogeny_fp)
            filepaths.append((phylogeny_fp, 'plain_text'))
        except Exception:
            return False, None, ("Phylogenetic tree cannot be parsed "
                                 "via scikit-biom")

    for fp_type, fps in files.items():
        if fp_type not in ('biom', 'preprocessed_fasta', 'plain_text'):
            for fp in fps:
                filepaths.append((fp, fp_type))

    index_fp, viz_fp, qza_fp = _generate_html_summary(
        new_biom_fp, md, join(out_dir), is_analysis, tree)

    filepaths.append((index_fp, 'html_summary'))
    filepaths.append((viz_fp, 'html_summary_dir'))
    if 'qza' not in files:
        filepaths.append((qza_fp, 'qza'))

    return True, [ArtifactInfo(None, 'BIOM', filepaths)], ""

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testFullTaxonomy(self):
     ann = TreeAnnotator()
     tree = TreeNode.read(StringIO("(((A:1, B:2):3, (C:4, D:5)'g__genus2':6)'f__family':10)root;"))
     assert_equals('f__family; g__genus2', ann.full_taxonomy(tree, tree.find('D')))

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def testSistersOneIncompleteSister(self):
     ann = TreeAnnotator()
     tree = TreeNode.read(StringIO("((A:1, B:2):3, ((C:1,D:1):1, (E:1,F:5)'g3':6):10)root;"))
     print(tree.ascii_art())
     sisters = ann.find_sisters(tree, tree.find('B'))
     assert_equals(sorted(['g3']), sorted([s.name for s in sisters]))

# 需要导入模块: from skbio.tree import TreeNode [as 别名]
# 或者: from skbio.tree.TreeNode import read [as 别名]
 def test_missing_taxonomy(self):
     tree = TreeNode.read(StringIO('((((A:11, B:12)C:10, D:9)E:20, F:20)G:30)root;'))
     assert_equals(['C'], TaxonomyFunctions().missing_taxonomy(tree, tree.find('A'), tree.find('E')))
     assert_equals([], TaxonomyFunctions().missing_taxonomy(tree, tree.find('A'), tree.find('A')))
     assert_equals(['E','C'], TaxonomyFunctions().missing_taxonomy(tree, tree.find('A'), tree.find('G')))