-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Refactoring to fix network related bugs
- Loading branch information
Showing
42 changed files
with
885 additions
and
764 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file was deleted.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,106 @@ | ||
| """Graph Util Class.""" | ||
|
|
||
| from hdash.synapse.meta_map import MetaMap | ||
| from hdash.util.id_util import IdUtil | ||
| from hdash.validator.categories import Categories | ||
| from hdash.graph.node_data import NodeData | ||
| from hdash.graph.htan_graph import HtanGraph | ||
|
|
||
|
|
||
| class GraphCreator: | ||
| """ | ||
| Creates a Graph of Atlas Nodes. | ||
| Given a set of MetaFiles, create an HTAN Graph. | ||
| This enables us to link patients --> biospecimens --> assays. | ||
| """ | ||
|
|
||
| def __init__(self, atlas_id, meta_map: MetaMap): | ||
| """Default Constructor.""" | ||
| self._atlas_id = atlas_id | ||
| self._graph = HtanGraph() | ||
| self._meta_map = meta_map | ||
| self._categories = Categories() | ||
| self._id_util = IdUtil() | ||
| self.__gather_nodes() | ||
| self.__gather_edges() | ||
|
|
||
| @property | ||
| def htan_graph(self): | ||
| return self._graph | ||
|
|
||
| def __gather_nodes(self): | ||
| """Gather all nodes.""" | ||
| self.__gather_nodes_by_category(self._categories.DEMOGRAPHICS) | ||
| self.__gather_nodes_by_category(self._categories.BIOSPECIMEN) | ||
| self.__gather_nodes_by_category(self._categories.SRRS_BIOSPECIMEN) | ||
| for category in self._categories.all_assays: | ||
| self.__gather_nodes_by_category(category) | ||
|
|
||
| def __gather_nodes_by_category(self, category): | ||
| """Gather all Nodes in the Specified Category.""" | ||
| meta_file_list = self._meta_map.get_meta_file_list(category) | ||
| for meta_file in meta_file_list: | ||
| df = meta_file.df | ||
| primary_id = self._id_util.get_primary_id_column(category) | ||
| id_list = df[primary_id].to_list() | ||
|
|
||
| # Each Primary ID Gets its Own Node | ||
| for current_id in id_list: | ||
| current_id = str(current_id) | ||
| node_data = NodeData(current_id, meta_file) | ||
| self._graph.add_node(node_data) | ||
|
|
||
| def __gather_edges(self): | ||
| """Gather all the edges.""" | ||
| for category in self._categories.all_categories: | ||
| self.__gather_edges_by_category(category) | ||
|
|
||
| def __gather_edges_by_category(self, category): | ||
| meta_file_list = self._meta_map.get_meta_file_list(category) | ||
| for meta_file in meta_file_list: | ||
| df = meta_file.df | ||
| primary_id_col = self._id_util.get_primary_id_column(category) | ||
| parent_id_col = self._id_util.get_parent_id_column(category) | ||
| adj_id_col = self._id_util.get_adjacent_id_column(category) | ||
| if parent_id_col is not None: | ||
| self.__gather_child_parent_edges(df, primary_id_col, parent_id_col) | ||
| if adj_id_col is not None and adj_id_col in df.columns: | ||
| self.__gather_adjacent_edges(df, primary_id_col, adj_id_col) | ||
|
|
||
| def __gather_child_parent_edges(self, df, primary_id_col, parent_id_col): | ||
| """Gather Parent Child Edges.""" | ||
| for index, row in df.iterrows(): | ||
| primary_id = str(row[primary_id_col]) | ||
| parent_id_chunk = str(row[parent_id_col]) | ||
| parent_id_chunk = self.__handle_htapp_special_case(parent_id_chunk, row) | ||
|
|
||
| # We can have multiple parents | ||
| parent_id_chunk = parent_id_chunk.replace(";", " ").replace(",", " ") | ||
| parts = parent_id_chunk.split() | ||
| for part in parts: | ||
| parent_id = part.strip() | ||
| self._graph.add_edge(parent_id, primary_id) | ||
|
|
||
| def __handle_htapp_special_case(self, parent_id_chunk, row): | ||
| """Special case handling for HTAPP/DFCI.""" | ||
| if parent_id_chunk.startswith("Not Applicable"): | ||
| try: | ||
| parent_id_chunk = str(row[IdUtil.HTAN_PARENT_BIOSPECIMEN_ID]) | ||
| except KeyError: | ||
| parent_id_chunk = "NOT_APPLICABLE" | ||
| return parent_id_chunk | ||
|
|
||
| def __gather_adjacent_edges(self, df, primary_id_col, adj_id_col): | ||
| """Gather Adjacent Edges.""" | ||
| for index, row in df.iterrows(): | ||
| adj_id_chunk = str(row[adj_id_col]) | ||
| primary_id = str(row[primary_id_col]) | ||
|
|
||
| # We can have multiple adjacent nodes | ||
| if adj_id_chunk != "nan": | ||
| adj_id_chunk = adj_id_chunk.replace(";", " ").replace(",", " ") | ||
| parts = adj_id_chunk.split() | ||
| for part in parts: | ||
| adjacent_id = part.strip() | ||
| self._graph.add_adjacency_edge(primary_id, adjacent_id) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,80 @@ | ||
| """Graph Flattener Class.""" | ||
| import networkx as nx | ||
| from natsort import natsorted | ||
| from hdash.graph.htan_graph import HtanGraph | ||
| from hdash.validator.categories import Categories | ||
| from hdash.graph.key import KeyUtil | ||
|
|
||
|
|
||
| class GraphFlattener: | ||
| """Graph Flattener Class. | ||
| Given an HTAN Graph, flatten it so that we can map: | ||
| 1) Participant --> All Derived Biospecimens. | ||
| 2) Biospecimen --> All Derived Assays. | ||
| """ | ||
|
|
||
| def __init__(self, htan_graph: HtanGraph): | ||
| """Default Constructor.""" | ||
| self.htan_graph = htan_graph | ||
| self.directed_graph = htan_graph.directed_graph | ||
| self.categories = Categories() | ||
| self.participant_2_biopsecimens = {} | ||
| self.biospecimen_2_assays = {} | ||
| self.assay_map = set() | ||
| self.__bin_nodes() | ||
| self.__gather_downstream_biospecimens() | ||
| self.__gather_downstream_assays() | ||
|
|
||
| def biospecimen_has_assay(self, biospecimen_id, category): | ||
| """Determine if the specified biospecimen has the specified assay.""" | ||
| key = KeyUtil.create_key(biospecimen_id, category) | ||
| return key in self.assay_map | ||
|
|
||
| def __bin_nodes(self): | ||
| """Bin Participants and Biospecimens.""" | ||
| self.participant_id_set = set() | ||
| self.biospecimen_id_set = set() | ||
| for node_id in self.directed_graph.nodes: | ||
| data = self.directed_graph.nodes[node_id][HtanGraph.DATA_KEY] | ||
| category = data.meta_file.category | ||
| if category == self.categories.DEMOGRAPHICS: | ||
| self.participant_id_set.add(node_id) | ||
| elif category in self.categories.biospecimen_list: | ||
| self.biospecimen_id_set.add(node_id) | ||
|
|
||
| # Sort the Participants | ||
| self.participant_id_set = natsorted(self.participant_id_set) | ||
|
|
||
| def __gather_downstream_biospecimens(self): | ||
| """Given a Participant, find *all* Downstream Biospecimens.""" | ||
| for participant_id in self.participant_id_set: | ||
| nodes = nx.descendants(self.directed_graph, participant_id) | ||
|
|
||
| # Filter Descendents for Biospecimens Only | ||
| filtered_list = self.__filter_nodes(nodes, self.categories.biospecimen_list) | ||
| self.participant_2_biopsecimens[participant_id] = filtered_list | ||
|
|
||
| def __gather_downstream_assays(self): | ||
| """Given a Biospecimen, find *all* Downstream Assays.""" | ||
| for biospecimen_id in self.biospecimen_id_set: | ||
| nodes = nx.descendants(self.directed_graph, biospecimen_id) | ||
|
|
||
| # Filter Descendents for Assays Only | ||
| filtered_list = self.__filter_nodes(nodes, self.categories.all_assays) | ||
| self.biospecimen_2_assays[biospecimen_id] = filtered_list | ||
|
|
||
| # Add to assay map for easy look-up | ||
| for node_id in filtered_list: | ||
| data = self.directed_graph.nodes[node_id][HtanGraph.DATA_KEY] | ||
| key = KeyUtil.create_key(biospecimen_id, data.meta_file.category) | ||
| self.assay_map.add(key) | ||
|
|
||
| def __filter_nodes(self, nodes, target_categories): | ||
| """Filter Node List to Only those in the Target Categories.""" | ||
| filtered_list = [] | ||
| for node_id in nodes: | ||
| data = self.directed_graph.nodes[node_id][HtanGraph.DATA_KEY] | ||
| if data.meta_file.category in target_categories: | ||
| filtered_list.append(node_id) | ||
| return filtered_list |
Oops, something went wrong.