Pangraph 0.3.0

anvio/__init__.py

@@ -181,6 +181,14 @@ def TABULATE(table, header, numalign="right", max_width=0):
                      "You can provide these names as a comma-separated list of names, or you can put them in a file, "
                      "where you have a single genome name in each line, and provide the file path."}
                 ),
+    'genomes-reverse': (
+            ['-R', '--genome-reverse'],
+            {'metavar': "GENOME_NAMES",
+             'required': False,
+             'help': "Genome names to 'reverse'. You can use this parameter to choose the genomes whose synteny you want to reverse. "
+                     "You can provide these names as a comma-separated list of names, or you can put them in a file, "
+                     "where you have a single genome name in each line, and provide the file path."}
+                ),
     'blank-profile': (
             ['--blank-profile'],
             {'default': False,
@@ -2115,6 +2123,74 @@ def TABULATE(table, header, numalign="right", max_width=0):
              'type': str,
              'help': "Directory path for output files"}
                 ),
+     # start of anvi-pan-graph related input and output variables:
+    'testing-yaml': (
+            ['--testing-yaml'],
+            {'metavar': 'YAML',
+             'type': str,
+             'help': "A yaml file containing raw gene cluster order for testing the anvi'o pan graph."}
+                ),
+    'gene-additional-data': (
+            ['--gene-additional-data'],
+            {'metavar': 'CSV',
+             'type': str,
+             'help': "A CSV file containing at least one column with values per genome, contig and genecall column."}
+                ),
+    'gc-additional-data': (
+            ['--gc-additional-data'],
+            {'metavar': 'CSV',
+             'type': str,
+             'help': "A CSV file containing at least one column with values per genome, contig and gc column."}
+                ),
+     'output-pan-graph-json': (
+            ['-o', '--output-pan-graph-json'],
+            {'metavar': 'FILE_PATH',
+             'type': str,
+             'help': "File path to store the temporary pan-graph-json file."}
+                ),
+     'output-graphml': (
+            ['--output-graphml'],
+            {'metavar': 'FILE_PATH',
+             'type': str,
+             'help': "File path to store the final graph file."}
+                ),
+     'output-raw-gc-additional-data': (
+            ['--output-raw-gc-additional-data'],
+            {'metavar': 'FILE_PATH',
+             'type': str,
+             'help': "File path to create and store a raw gc additional data table for testing or join purposes."}
+                ),
+     'output-raw-gene-additional-data': (
+            ['--output-raw-gene-additional-data'],
+            {'metavar': 'FILE_PATH',
+             'type': str,
+             'help': "File path to create and store a raw gc additional data table for testing or join purposes."}
+                ),
+     'output-testing-yaml': (
+            ['--output-testing-yaml'],
+            {'metavar': 'FILE_PATH',
+             'type': str,
+             'help': "File path to create and store a testing yaml for a existing pangenome."}
+                ),
+     'output-dir-summary': (
+            ['--output-dir-summary'],
+            {'metavar': 'DIR_PATH',
+             'type': str,
+             'help': "File path to store a CSV file containing values describing the pan-graph for downstream analysis."}
+                ),
+     'output-dir-graphics': (
+            ['--output-dir-graphics'],
+            {'metavar': 'DIR_PATH',
+             'type': str,
+             'help': "File path to store the hierarchical clusterings for the paralog splitting."}
+                ),
+     'pan-graph-json': (
+            ['-i', '--pan-graph-json'],
+            {'metavar': 'JSON',
+             'type': str,
+             'help': "A JSON formatted anvi'o pan graph, which is typically generated by `anvi-pan-graph`."}
+                ),
+    # end of pan-graph related variables
     'output-file': (
             ['-o', '--output-file'],
             {'metavar': 'FILE_PATH',
@@ -3701,6 +3777,7 @@ def set_version():
            anvio_codename, \
            t.contigs_db_version, \
            t.pan_db_version, \
+           t.pangraph_json_version, \
            t.profile_db_version, \
            t.genes_db_version, \
            t.auxiliary_data_version, \
@@ -3717,6 +3794,7 @@ def get_version_tuples():
             ("Genes DB version", __genes__version__),
             ("Auxiliary data storage version", __auxiliary_data_version__),
             ("Pan DB version", __pan__version__),
+            ("Pangraph JSON version", __pangraph__version__),
             ("Genome data storage version", __genomes_storage_version__),
             ("Structure DB version", __structure__version__),
             ("KEGG Modules DB version", __kegg_modules_version__),
@@ -3729,6 +3807,7 @@ def print_version():
     run.info("Profile database", __profile__version__)
     run.info("Contigs database", __contigs__version__)
     run.info("Pan database", __pan__version__)
+    run.info("Pangraph JSON", __pangraph__version__)
     run.info("Genome data storage", __genomes_storage_version__)
     run.info("Auxiliary data storage", __auxiliary_data_version__)
     run.info("Structure database", __structure__version__)
@@ -3740,6 +3819,7 @@ def print_version():
 __codename__, \
 __contigs__version__, \
 __pan__version__, \
+__pangraph__version__, \
 __profile__version__, \
 __genes__version__, \
 __auxiliary_data_version__, \

anvio/bottleroutes.py

@@ -21,6 +21,7 @@
 import argparse
 import datetime
 import importlib
+import networkx as nx
 
 from hashlib import md5
 from ete3 import Tree
@@ -38,6 +39,7 @@
 
 import anvio
 import anvio.dbops as dbops
+import anvio.panops as panops
 import anvio.utils as utils
 import anvio.drivers as drivers
 import anvio.terminal as terminal
@@ -197,7 +199,12 @@ def register_routes(self):
         self.route('/data/get_gene_info/<gene_callers_id>',    callback=self.get_gene_info)
         self.route('/data/get_metabolism',                     callback=self.get_metabolism)
         self.route('/data/get_scale_bar',                      callback=self.get_scale_bar, method='POST')
-
+        self.route('/pangraph/settings',                       callback=self.get_pangraph_settings, method="POST")
+        self.route('/pangraph/get_json',                       callback=self.get_pangraph_json_data, method="POST")
+        self.route('/pangraph/alignment',                      callback=self.get_pangraph_gc_alignment, method="POST")
+        self.route('/pangraph/analyse',                        callback=self.get_pangraph_bin_summary, method="POST")
+        self.route('/pangraph/function',                       callback=self.get_pangraph_gc_function, method="POST")
+        self.route('/pangraph/filter',                         callback=self.get_pangraph_passed_gene_clusters, method="POST")
 
     def run_application(self, ip, port):
         # check for the wsgi module bottle will use.
@@ -266,6 +273,8 @@ def redirect_to_app(self):
             homepage = 'structure.html'
         elif self.interactive.mode == 'metabolism':
             homepage = 'metabolism.html'
+        elif self.interactive.mode == 'pangraph':
+            homepage = 'pangraph.html'
         elif self.interactive.mode == 'inspect':
             redirect('/app/charts.html?id=%s&show_snvs=true&rand=%s' % (self.interactive.inspect_split_name, self.random_hash(8)))
 
@@ -1069,7 +1078,7 @@ def gen_summary(self, collection_name):
 
         path = "summary/%s/index.html" % (collection_name)
         return json.dumps({'path': path})
-        
+
 
     def send_summary_static(self, collection_name, filename):
         if self.interactive.mode == 'pan':
@@ -1534,7 +1543,7 @@ def get_functions_for_gene_clusters(self):
                 message = (f"At least one of the gene clusters in your list (e.g., {gene_cluster_name}) is missing in "
                            f"the functions summary dict :/")
                 return json.dumps({'status': 1, 'message': message})
-                
+
             d[gene_cluster_name] = self.interactive.gene_clusters_functions_summary_dict[gene_cluster_name]
 
         return json.dumps({'functions': d, 'sources': list(self.interactive.gene_clusters_function_sources)})
@@ -1551,4 +1560,133 @@ def get_scale_bar(self):
             message = str(e.clear_text()) if hasattr(e, 'clear_text') else str(e)
             return json.dumps({'status': 1, 'message': message})
 
-        return json.dumps({'scale_bar_value': total_branch_length})
+        return json.dumps({'scale_bar_value': total_branch_length})
+
+    def get_pangraph_json_data(self):
+
+        return self.interactive.get_pangraph_json()
+
+
+    def get_pangraph_bin_summary(self):
+
+        payload = request.json
+        selection = payload['selection']
+        result = {}
+
+        self.interactive.init_gene_clusters()
+        self.interactive.init_gene_clusters_functions_summary_dict()
+
+        bin_functions = {}
+        all_functions = []
+
+        for bin in list(payload.keys())[1:]:
+
+            bin_functions[bin] = []
+            gene_cluster_list = payload[bin]
+            for gene_cluster_name in gene_cluster_list:
+                function = self.interactive.gene_clusters_functions_summary_dict[gene_cluster_name][selection]['function']
+
+                bin_functions[bin].append(function)
+                all_functions.append(function)
+
+        for function in all_functions:
+            result[function] = {}
+            for bin in bin_functions.keys():
+                if function in bin_functions[bin]:
+                    result[function][bin] = 1
+                else:
+                    result[function][bin] = 0
+
+        return json.dumps(result)
+
+    def get_pangraph_settings(self):
+
+        payload = request.json
+        max_edge_length_filter = payload['maxlength']
+        gene_cluster_grouping_threshold = payload['condtr']
+        groupcompress = payload['groupcompress']
+        ungrouping_area = payload['ungroup']
+        print(ungrouping_area)
+
+        args = argparse.Namespace(
+            pan_graph_json=self.interactive.pan_graph_json_path,
+            output_pan_graph_json=self.interactive.pan_graph_json_path,
+            max_edge_length_filter=max_edge_length_filter,
+            gene_cluster_grouping_threshold=gene_cluster_grouping_threshold,
+            gene_cluster_grouping_compression=groupcompress,
+            ungrouping_area = ungrouping_area
+            )
+
+        Pangraph = panops.Pangraph(args)
+
+        Pangraph.load_graph_from_json_file()
+        Pangraph.run_synteny_layout_algorithm()
+        Pangraph.update_json_dict()
+        Pangraph.store_network()
+
+        return({'status': 0})
+
+    def get_pangraph_gc_alignment(self):
+
+        payload = request.json
+        result = {}
+
+        for genome in payload.keys():
+            genecall, contig, direction, name = payload[genome]
+            gene_cluster_alignment_dict = self.interactive.get_sequences_for_gene_clusters(gene_cluster_names=set([name]), skip_alignments=False, report_DNA_sequences=False)[name]
+            sequence = gene_cluster_alignment_dict[genome][int(genecall)]
+
+            result[genome] = [genecall, contig, direction, sequence]
+
+        return(result)
+
+    def get_pangraph_gc_function(self):
+
+        payload = request.json
+        result = {}
+
+        self.interactive.init_gene_clusters()
+        self.interactive.init_gene_clusters_functions_summary_dict()
+
+        gene_cluster_name = payload['genecluster']
+
+        for selection in self.interactive.gene_clusters_functions_summary_dict[gene_cluster_name].keys():
+            result[selection] = self.interactive.gene_clusters_functions_summary_dict[gene_cluster_name][selection]
+        return json.dumps(result)
+
+    def get_pangraph_passed_gene_clusters(self):
+
+        payload = request.json
+        result = {'gene_clusters': []}
+
+        self.interactive.init_gene_clusters()
+        self.interactive.init_gene_clusters_functions_summary_dict()
+
+        functions_dict = self.interactive.gene_clusters_functions_summary_dict
+
+        for gene_cluster in functions_dict.keys():
+            for selection in payload.keys():
+                if selection in functions_dict[gene_cluster].keys():
+                    for search in payload[selection]:
+                        value = functions_dict[gene_cluster][selection]
+
+                        accession = value['accession']
+                        func = value['function']
+
+                        if not accession or not func:
+                            accession = ''
+                            func = ''
+
+                        if search == '':
+
+                            print(search, accession, func)
+                            if search == accession or search == func:
+                                if gene_cluster not in result['gene_clusters']:
+                                    result['gene_clusters'] += [gene_cluster]
+
+                        else:
+                            if search.lower() in accession.lower() or search.lower() in func.lower():
+                                if gene_cluster not in result['gene_clusters']:
+                                    result['gene_clusters'] += [gene_cluster]
+
+        return json.dumps(result)

anvio/clustering.py

@@ -103,6 +103,23 @@ def get_newick_tree_data_for_dict(d, transpose=False, linkage=constants.linkage_
     return newick
 
 
+def get_newick(node, parent_dist, leaf_names, newick=''):
+    """
+    Modified from the solution at https://stackoverflow.com/questions/28222179/save-dendrogram-to-newick-format
+    """
+    if node.is_leaf():
+        return "%s:%.2f%s" % (leaf_names[node.id], parent_dist - node.dist, newick)
+    else:
+        if len(newick) > 0:
+            newick = "):%.2f%s" % (parent_dist - node.dist, newick)
+        else:
+            newick = ");"
+        newick = get_newick(node.get_left(), node.dist, leaf_names, newick=newick)
+        newick = get_newick(node.get_right(), node.dist, leaf_names, newick=",%s" % (newick))
+        newick = "(%s" % (newick)
+        return newick
+
+
 def get_vectors_for_vectors_with_missing_data(vectors):
     """Get a distance matrix for vectors with missing data.