biocypher 0.9.7__py3-none-any.whl → 0.10.1__py3-none-any.whl

biocypher/_metadata.py

@@ -10,7 +10,7 @@ import pathlib
 
 import toml
 
-_VERSION = "0.9.7"
+_VERSION = "0.10.1"
 
 
 def get_metadata():

biocypher/output/in_memory/_airr.py

@@ -0,0 +1,499 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any, Optional
+
+from biocypher._create import BioCypherEdge, BioCypherNode
+from biocypher._deduplicate import Deduplicator
+from biocypher._logger import logger
+from biocypher.output.in_memory._in_memory_kg import _InMemoryKG
+
+if TYPE_CHECKING:
+    from scirpy.io import AirrCell
+
+try:
+    from scirpy.io import AirrCell
+
+    HAS_SCIRPY = True
+except ImportError:
+    HAS_SCIRPY = False
+
+
+class AirrKG(_InMemoryKG):
+    """Knowledge graph for AIRR (Adaptive Immune Receptor Repertoire) data.
+
+    This class implements the AIRR data model for representing immune receptor sequences
+    (antibodies and T cell receptors) and their annotations. To ensure proper conversion
+    to AIRR format, your schema file should define immune receptor entities with property
+    names that match the AIRR standards.
+
+    Key property names in your schema for immune receptor entities:
+    - locus: The gene locus (e.g., "TRA", "TRB", "IGH", "IGK", "IGL")
+    - junction_aa: The amino acid sequence of the junction region (CDR3)
+    - v_call: The V gene assignment
+    - j_call: The J gene assignment
+    - productive: Whether the sequence is productive
+
+    Pairing Strategies specified in get_kg method:
+    - Indirect pairings allowed:
+        Epitope is only matched with ONE of the paired receptors -> the "paired" AIRR cell will be created
+    - Indirect pairings not allowed:
+        Epitope is only matched with ONE of the paired receptors -> no "paired" AIRR cell will be created
+
+    For a complete list of available fields and their descriptions, see:
+    https://docs.airr-community.org/en/stable/datarep/rearrangements.html#fields
+
+    All properties from the biocypher schema defined by user will be preserved in the AIRR format.
+    """
+
+    # Constants for internal property filtering
+    _INTERNAL_PROPERTIES = frozenset(["node_id", "node_label", "id", "preferred_id"])
+
+    def __init__(
+        self,
+        deduplicator: Optional["Deduplicator"] = None,
+        metadata_entity_type: str = "epitope",
+    ) -> None:
+        """Initialize AirrKG with configurable metadata node type.
+
+        Args:
+        ----
+            deduplicator: Deduplicator instance
+            metadata_entity_type: String specifying the metadata node type (default: "epitope")
+
+        """
+        super().__init__()
+        self.deduplicator = deduplicator or Deduplicator()
+        self.metadata_entity_type = metadata_entity_type
+
+        # Initialize storage for processed cells
+        self.adjacency_list = {}
+        self.airr_cells = []
+
+        # These will be populated when nodes and edges are added
+        self.sequence_entity_types = {}
+        self.chain_relationship_types = []
+        self.chain_to_epitope_relationship_types = []
+
+    def _check_dependencies(self) -> None:
+        """Verify that scirpy is available."""
+        if not HAS_SCIRPY:
+            msg = (
+                "AirrCell module from scirpy not detected. "
+                "Install it with 'poetry add biocypher[scirpy]' or 'poetry add scirpy'."
+            )
+            raise ImportError(msg)
+
+    def get_kg(self, indirect_pairings: bool = True) -> list[AirrCell]:
+        """Convert directly to AIRR format using AirCell from scirpy.
+
+        Args:
+        ----
+            indirect_pairings: Boolean controlling pairing strategy (default: True)
+                - True:
+                    Epitope is only matched with ONE of the paired receptors -> the "paired" AIRR cell will be created
+                - False:
+                    Epitope is only matched with ONE of the paired receptors -> no "paired" AIRR cell will be created
+
+        Returns:
+        -------
+            list: List of generated AIRR cells
+
+        """
+        self._check_dependencies()
+        if not self.airr_cells:
+            self.airr_cells = self._to_airr_cells(self.adjacency_list, indirect_pairings)
+        return self.airr_cells
+
+    def add_nodes(self, nodes: list[BioCypherNode]) -> None:
+        """Add BioCypher nodes, organizing them by type."""
+        self._add_to_entities_by_type(nodes)
+
+    def add_edges(self, edges: list[BioCypherEdge]) -> None:
+        """Add BioCypher edges, organizing them by type."""
+        self._add_to_entities_by_type(edges)
+
+    def _add_to_entities_by_type(self, entities: dict[str, list[Any]]) -> None:
+        """Add all entities (both nodes and edges) to a common adj. list."""
+        lists = self._separate_entity_types(entities)
+        for _type, _entities in lists.items():
+            if _type not in self.adjacency_list:
+                self.adjacency_list[_type] = []
+            self.adjacency_list[_type].extend(_entities)
+
+    def _process_entities(self, entities: dict[str, list[Any]]) -> tuple[dict, dict, dict]:
+        """Process entities and organize them into sequence nodes, metadata nodes, and receptor-epitope mappings.
+
+        Args:
+        ----
+            entities: Dictionary mapping entity types to lists of BioCypherNode/BioCypherEdge objects
+
+        Returns:
+        -------
+            tuple: (sequence_nodes, metadata_nodes, receptor_epitope_mapping)
+
+        """
+        sequence_nodes = {}
+        metadata_nodes = {}
+        receptor_epitope_mapping = {}
+
+        # Determine entity types while processing
+        all_node_types = set()
+        all_edge_types = set()
+
+        for entity_type, entities_list in entities.items():
+            if not entities_list:  # Skip empty lists
+                continue
+
+            # Determine if this is a node or edge type
+            if isinstance(entities_list[0], BioCypherNode):
+                all_node_types.add(entity_type)
+                if entity_type == self.metadata_entity_type:
+                    metadata_nodes.update({node.get_id(): node for node in entities_list})
+                else:
+                    sequence_nodes.update({node.get_id(): node for node in entities_list})
+                    self.sequence_entity_types[entity_type] = entity_type.replace(" sequence", "").upper()
+            elif isinstance(entities_list[0], BioCypherEdge):
+                all_edge_types.add(entity_type)
+
+        # Update relationship types
+        self.chain_relationship_types = [
+            edge_type for edge_type in all_edge_types if self.metadata_entity_type not in edge_type.lower()
+        ]
+
+        self.chain_to_epitope_relationship_types = [
+            edge_type for edge_type in all_edge_types if self.metadata_entity_type in edge_type.lower()
+        ]
+
+        # Process chain-to-epitope relationships
+        for entity_type in self.chain_to_epitope_relationship_types:
+            self._update_receptor_epitope_mapping(entities[entity_type], receptor_epitope_mapping)
+
+        return sequence_nodes, metadata_nodes, receptor_epitope_mapping
+
+    def _update_receptor_epitope_mapping(self, edges: list[BioCypherEdge], mapping: dict[str, set]) -> None:
+        """Update receptor-epitope mapping with new edges.
+
+        Args:
+        ----
+            edges: List of edges to process
+            mapping: Dictionary to update with receptor-epitope mappings
+
+        """
+        for edge in edges:
+            source_id = edge.get_source_id()
+            if source_id not in mapping:
+                mapping[source_id] = set()
+            mapping[source_id].add(edge.get_target_id())
+
+    def _process_paired_chains(
+        self,
+        entities: dict[str, list[Any]],
+        sequence_nodes: dict[str, BioCypherNode],
+        metadata_nodes: dict[str, BioCypherNode],
+        receptor_epitope_mapping: dict[str, set],
+        indirect_pairings: bool = True,
+    ) -> tuple[list[AirrCell], set[str], int]:
+        """Process paired chains and generate AIRR cells.
+
+        Args:
+        ----
+            entities: Dictionary of all entities
+            sequence_nodes: Dictionary of sequence nodes
+            metadata_nodes: Dictionary of metadata nodes
+            receptor_epitope_mapping: Dictionary of receptor-epitope mappings
+            indirect_pairings: Boolean controlling pairing strategy
+
+        Returns:
+        -------
+            tuple: (list of generated cells, set of processed chain IDs, count of cells with multiple epitopes)
+
+        """
+        airr_cells = []
+        processed_chains = set()
+        n_metacells = 0
+
+        for entity_type, edges in entities.items():
+            if entity_type in self.chain_relationship_types:
+                for edge in edges:
+                    source_id, target_id = edge.get_source_id(), edge.get_target_id()
+                    processed_chains.update([source_id, target_id])
+
+                    # Use conditional logic for pairing strategy
+                    source_metadata = receptor_epitope_mapping.get(source_id, set())
+                    target_metadata = receptor_epitope_mapping.get(target_id, set())
+
+                    if indirect_pairings:
+                        # Union: create paired cell if either chain binds epitopes
+                        metadata_ids = source_metadata | target_metadata
+
+                        metadata_nodes_cell = self._get_metadata_nodes(metadata_ids, metadata_nodes)
+                        if metadata_nodes_cell:
+                            cell_s = self._generate_airr_cell(
+                                cell_id=edge.get_id(),
+                                source_node=sequence_nodes.get(source_id),
+                                target_node=sequence_nodes.get(target_id),
+                                metadata_nodes=metadata_nodes_cell,
+                                paired=True,
+                                receptor_epitope_mapping=receptor_epitope_mapping,
+                            )
+                            airr_cells.extend(cell_s)
+                            if len(cell_s) > 1:
+                                n_metacells += 1
+                    else:
+                        # Intersection: create paired cell only if both chains bind same epitopes
+                        shared_metadata_ids = source_metadata & target_metadata
+
+                        # Create paired cell if there are shared epitopes
+                        if shared_metadata_ids:
+                            shared_metadata_nodes = self._get_metadata_nodes(shared_metadata_ids, metadata_nodes)
+                            if shared_metadata_nodes:
+                                cell_s = self._generate_airr_cell(
+                                    cell_id=edge.get_id(),
+                                    source_node=sequence_nodes.get(source_id),
+                                    target_node=sequence_nodes.get(target_id),
+                                    metadata_nodes=shared_metadata_nodes,
+                                    paired=True,
+                                    receptor_epitope_mapping=receptor_epitope_mapping,
+                                )
+                                airr_cells.extend(cell_s)
+                                if len(cell_s) > 1:
+                                    n_metacells += 1
+
+                        # Create unpaired cells for chains with non-overlapping epitopes
+                        source_only_metadata = source_metadata - target_metadata
+                        target_only_metadata = target_metadata - source_metadata
+
+                        # Create unpaired cell for source chain if it has unique epitopes
+                        if source_only_metadata:
+                            source_only_nodes = self._get_metadata_nodes(source_only_metadata, metadata_nodes)
+                            if source_only_nodes:
+                                source_cells = self._generate_airr_cell(
+                                    cell_id=f"unpaired_{source_id}",
+                                    source_node=sequence_nodes.get(source_id),
+                                    target_node=None,
+                                    metadata_nodes=source_only_nodes,
+                                    paired=False,
+                                    receptor_epitope_mapping=receptor_epitope_mapping,
+                                )
+                                airr_cells.extend(source_cells)
+                                if len(source_cells) > 1:
+                                    n_metacells += 1
+
+                        # Create unpaired cell for target chain if it has unique epitopes
+                        if target_only_metadata:
+                            target_only_nodes = self._get_metadata_nodes(target_only_metadata, metadata_nodes)
+                            if target_only_nodes:
+                                target_cells = self._generate_airr_cell(
+                                    cell_id=f"unpaired_{target_id}",
+                                    source_node=sequence_nodes.get(target_id),
+                                    target_node=None,
+                                    metadata_nodes=target_only_nodes,
+                                    paired=False,
+                                    receptor_epitope_mapping=receptor_epitope_mapping,
+                                )
+                                airr_cells.extend(target_cells)
+                                if len(target_cells) > 1:
+                                    n_metacells += 1
+
+        return airr_cells, processed_chains, n_metacells
+
+    def _process_unpaired_chains(
+        self,
+        receptor_epitope_mapping: dict[str, set],
+        sequence_nodes: dict[str, BioCypherNode],
+        metadata_nodes: dict[str, BioCypherNode],
+        processed_chains: set[str],
+    ) -> tuple[list[AirrCell], int]:
+        """Process unpaired chains and generate AIRR cells.
+
+        Args:
+        ----
+            receptor_epitope_mapping: Dictionary of receptor-epitope mappings
+            sequence_nodes: Dictionary of sequence nodes
+            metadata_nodes: Dictionary of metadata nodes
+            processed_chains: Set of already processed chain IDs
+
+        Returns:
+        -------
+            tuple: (List of generated cells, count of cells with multiple epitopes)
+
+        """
+        airr_cells = []
+        n_metacells = 0
+
+        for chain_id in receptor_epitope_mapping:
+            if chain_id not in processed_chains:
+                # Get all metadata nodes for this unpaired chain
+                metadata_nodes_cell = self._get_metadata_nodes(receptor_epitope_mapping[chain_id], metadata_nodes)
+
+                if metadata_nodes_cell:
+                    cell_s = self._generate_airr_cell(
+                        cell_id=f"unpaired_{chain_id}",
+                        source_node=sequence_nodes.get(chain_id),
+                        target_node=None,
+                        metadata_nodes=metadata_nodes_cell,
+                        paired=False,
+                        receptor_epitope_mapping=receptor_epitope_mapping,
+                    )
+                    airr_cells.extend(cell_s)
+                    # Check if multiple cells were generated (indicating multiple epitopes)
+                    if len(cell_s) > 1:
+                        n_metacells += 1
+
+        return airr_cells, n_metacells
+
+    def _to_airr_cells(self, entities: dict[str, list[Any]], indirect_pairings: bool = True) -> list[AirrCell]:
+        """Convert BioCypher entities to AIRR cells using configurable mappings.
+
+        Args:
+        ----
+            entities: Dictionary mapping entity types to lists of BioCypherNode/BioCypherEdge objects
+            indirect_pairings: Boolean controlling pairing strategy (default: True)
+                - True:
+                    Epitope is only matched with ONE of the paired receptors -> the "paired" AIRR cell will be created
+                - False:
+                    Epitope is only matched with ONE of the paired receptors -> no "paired" AIRR cell will be created
+
+        Returns:
+        -------
+            list: List of generated AIRR cells
+
+        """
+        self._check_dependencies()
+        if not entities:
+            msg = "No entities provided for conversion."
+            raise ValueError(msg)
+
+        logger.info("Starting conversion to AIRR cells")
+
+        # Process all entities
+        sequence_nodes, metadata_nodes, receptor_epitope_mapping = self._process_entities(entities)
+
+        # Process paired chains
+        airr_cells, processed_chains, paired_metacells = self._process_paired_chains(
+            entities,
+            sequence_nodes,
+            metadata_nodes,
+            receptor_epitope_mapping,
+            indirect_pairings,
+        )
+
+        # Process unpaired chains
+        unpaired_cells, unpaired_metacells = self._process_unpaired_chains(
+            receptor_epitope_mapping,
+            sequence_nodes,
+            metadata_nodes,
+            processed_chains,
+        )
+        airr_cells.extend(unpaired_cells)
+
+        # Calculate total cells with multiple epitopes
+        total_metacells = paired_metacells + unpaired_metacells
+
+        # Log information about cells
+        logger.info(f"Generated total of {len(airr_cells)} AIRR cells")
+        if total_metacells > 0:
+            logger.info(f"{total_metacells} cells with more than 1 epitope were detected")
+
+        return airr_cells
+
+    def _get_metadata_nodes(
+        self,
+        metadata_ids: set[str],
+        metadata_nodes: dict[str, BioCypherNode],
+    ) -> list[BioCypherNode]:
+        """Get metadata nodes for a set of metadata IDs.
+
+        Args:
+        ----
+            metadata_ids: Set of metadata IDs
+            metadata_nodes: Dictionary of metadata nodes
+
+        Returns:
+        -------
+            list: List of metadata nodes
+
+        """
+        return [metadata_nodes[ep_id] for ep_id in metadata_ids if ep_id in metadata_nodes]
+
+    def _generate_airr_cell(
+        self,
+        cell_id: str,
+        source_node: BioCypherNode | None,
+        target_node: BioCypherNode | None,
+        metadata_nodes: list[BioCypherNode],
+        paired: bool,
+        receptor_epitope_mapping: dict[str, set] | None = None,
+    ) -> list[AirrCell]:
+        self._check_dependencies()
+        cell = AirrCell(cell_id=cell_id)
+
+        # Process both chains
+        for node in [source_node, target_node]:
+            if not node:  # Skip if node is None
+                continue
+
+            props = node.get_properties()
+            chain = AirrCell.empty_chain_dict()
+
+            # Add all properties except internal ones
+            for key, value in props.items():
+                if key not in self._INTERNAL_PROPERTIES:
+                    chain[key] = value
+
+            # Add locus based on node type
+            chain["locus"] = self.sequence_entity_types.get(node.get_label(), node.get_label())
+            chain["consensus_count"] = 0
+            chain["productive"] = True
+
+            # Add binds_epitope field based on receptor_epitope_mapping
+            if receptor_epitope_mapping and node.get_id() in receptor_epitope_mapping:
+                chain["validated_epitope"] = bool(receptor_epitope_mapping[node.get_id()])
+            else:
+                chain["validated_epitope"] = False
+
+            cell.add_chain(chain)
+
+        # Add metadata
+        return self.add_metadata(metadata_nodes, cell, paired)
+
+    def add_metadata(self, metadata_nodes: list[BioCypherNode], cell: AirrCell, paired: bool) -> list[AirrCell]:
+        """Add metadata from nodes to cell(s) and return a list of cells.
+
+        Args:
+        ----
+            metadata_nodes: List of metadata nodes to add
+            cell: Base cell to add metadata to
+            paired: Whether the cell is paired
+
+        Returns:
+        -------
+            List of cells with metadata added
+
+        """
+        self._check_dependencies()
+        cells = []
+        if not metadata_nodes:
+            cell["data_source"] = "BioCypher"
+            cell["is_paired"] = paired
+            cells.append(cell)
+        else:
+            for i, node in enumerate(metadata_nodes):
+                # Create a new AirrCell for each metadata node
+                if i > 0:
+                    cell_id_new = f"{cell.cell_id}_meta{i+1}"
+                    meta_cell = AirrCell(cell_id=cell_id_new)
+                    for chain in cell.chains:
+                        meta_cell.add_chain(chain)
+                else:
+                    meta_cell = cell
+                props = node.get_properties()
+                for key, value in props.items():
+                    if key not in self._INTERNAL_PROPERTIES:
+                        meta_cell[key] = value
+
+                meta_cell["data_source"] = "BioCypher"
+                meta_cell["is_paired"] = paired
+
+                cells.append(meta_cell)
+        return cells

biocypher/output/in_memory/_get_in_memory_kg.py

@@ -8,6 +8,7 @@ from __future__ import annotations
 from typing import TYPE_CHECKING
 
 from biocypher._logger import logger
+from biocypher.output.in_memory._airr import AirrKG
 from biocypher.output.in_memory._networkx import NetworkxKG
 from biocypher.output.in_memory._pandas import PandasKG
 
@@ -19,7 +20,7 @@ logger.debug(f"Loading module {__name__}.")
 
 __all__ = ["get_in_memory_kg"]
 
-IN_MEMORY_DBMS = ["csv", "pandas", "tabular", "networkx"]
+IN_MEMORY_DBMS = ["csv", "pandas", "tabular", "networkx", "airr"]
 
 
 def get_in_memory_kg(
@@ -35,10 +36,11 @@ def get_in_memory_kg(
     """
     if dbms in ["csv", "pandas", "tabular"]:
         return PandasKG(deduplicator)
-
     if dbms == "networkx":
         return NetworkxKG(deduplicator)
-
-    msg = f"Getting the in memory BioCypher KG is not supported for the DBMS {dbms}. Supported: {IN_MEMORY_DBMS}."
-    logger.error(msg)
-    raise NotImplementedError(msg)
+    elif dbms == "airr":
+        return AirrKG(deduplicator)
+    else:
+        msg = f"Getting the in memory BioCypher KG is not supported for the DBMS {dbms}. Supported: {IN_MEMORY_DBMS}."
+        logger.error(msg)
+        raise NotImplementedError(msg)

biocypher/output/in_memory/_in_memory_kg.py

@@ -1,5 +1,7 @@
 from abc import ABC, abstractmethod
 
+from biocypher._create import BioCypherEdge, BioCypherNode, BioCypherRelAsNode
+
 
 class _InMemoryKG(ABC):
     """Abstract class for handling the in-memory Knowledge Graph instance.
@@ -9,6 +11,7 @@ class _InMemoryKG(ABC):
     - add_nodes
     - add_edges
     - get_kg
+    - _separate_entity_types
 
     Raises:
         NotImplementedError: InMemoryKG implementation must override 'add_nodes'
@@ -38,3 +41,57 @@ class _InMemoryKG(ABC):
     def get_kg(self):
         """Return the in-memory knowledge graph."""
         raise NotImplementedError("InMemoryKG implementation must override 'get_kg'")
+
+    def _separate_entity_types(self, entities):
+        """
+        Given mixed iterable of BioCypher objects, separate them into lists by
+        type. Also deduplicates using the `Deduplicator` instance.
+        """
+        lists = {}
+        for entity in entities:
+            if (
+                not isinstance(entity, BioCypherNode)
+                and not isinstance(entity, BioCypherEdge)
+                and not isinstance(entity, BioCypherRelAsNode)
+            ):
+                raise TypeError(
+                    "Expected a BioCypherNode / BioCypherEdge / " f"BioCypherRelAsNode, got {type(entity)}."
+                )
+
+            if isinstance(entity, BioCypherNode):
+                seen = self.deduplicator.node_seen(entity)
+            elif isinstance(entity, BioCypherEdge):
+                seen = self.deduplicator.edge_seen(entity)
+            elif isinstance(entity, BioCypherRelAsNode):
+                seen = self.deduplicator.rel_as_node_seen(entity)
+
+            if seen:
+                continue
+
+            if isinstance(entity, BioCypherRelAsNode):
+                node = entity.get_node()
+                source_edge = entity.get_source_edge()
+                target_edge = entity.get_target_edge()
+
+                _type = node.get_type()
+                if _type not in lists:
+                    lists[_type] = []
+                lists[_type].append(node)
+
+                _source_type = source_edge.get_type()
+                if _source_type not in lists:
+                    lists[_source_type] = []
+                lists[_source_type].append(source_edge)
+
+                _target_type = target_edge.get_type()
+                if _target_type not in lists:
+                    lists[_target_type] = []
+                lists[_target_type].append(target_edge)
+                continue
+
+            _type = entity.get_type()
+            if _type not in lists:
+                lists[_type] = []
+            lists[_type].append(entity)
+
+        return lists

biocypher/output/in_memory/_pandas.py

@@ -1,6 +1,5 @@
 import pandas as pd
 
-from biocypher._create import BioCypherEdge, BioCypherNode, BioCypherRelAsNode
 from biocypher.output.in_memory._in_memory_kg import _InMemoryKG
 
 
@@ -20,65 +19,8 @@ class PandasKG(_InMemoryKG):
     def add_edges(self, edges):
         self.add_tables(edges)
 
-    def _separate_entity_types(self, entities):
-        """
-        Given mixed iterable of BioCypher objects, separate them into lists by
-        type. Also deduplicates using the `Deduplicator` instance.
-        """
-        lists = {}
-        for entity in entities:
-            if (
-                not isinstance(entity, BioCypherNode)
-                and not isinstance(entity, BioCypherEdge)
-                and not isinstance(entity, BioCypherRelAsNode)
-            ):
-                raise TypeError(
-                    "Expected a BioCypherNode / BioCypherEdge / " f"BioCypherRelAsNode, got {type(entity)}."
-                )
-
-            if isinstance(entity, BioCypherNode):
-                seen = self.deduplicator.node_seen(entity)
-            elif isinstance(entity, BioCypherEdge):
-                seen = self.deduplicator.edge_seen(entity)
-            elif isinstance(entity, BioCypherRelAsNode):
-                seen = self.deduplicator.rel_as_node_seen(entity)
-
-            if seen:
-                continue
-
-            if isinstance(entity, BioCypherRelAsNode):
-                node = entity.get_node()
-                source_edge = entity.get_source_edge()
-                target_edge = entity.get_target_edge()
-
-                _type = node.get_type()
-                if _type not in lists:
-                    lists[_type] = []
-                lists[_type].append(node)
-
-                _source_type = source_edge.get_type()
-                if _source_type not in lists:
-                    lists[_source_type] = []
-                lists[_source_type].append(source_edge)
-
-                _target_type = target_edge.get_type()
-                if _target_type not in lists:
-                    lists[_target_type] = []
-                lists[_target_type].append(target_edge)
-                continue
-
-            _type = entity.get_type()
-            if _type not in lists:
-                lists[_type] = []
-            lists[_type].append(entity)
-
-        return lists
-
     def add_tables(self, entities):
-        """
-        Add Pandas dataframes for each node and edge type in the input.
-        """
-
+        """Add Pandas dataframes for each node and edge type in the input."""
         lists = self._separate_entity_types(entities)
 
         for _type, _entities in lists.items():

biocypher/output/write/_get_writer.py

@@ -9,6 +9,7 @@ from typing import TYPE_CHECKING
 from biocypher._config import config as _config
 from biocypher._logger import logger
 from biocypher.output.write._batch_writer import _BatchWriter
+from biocypher.output.write.graph._airr import _AirrWriter
 from biocypher.output.write.graph._arangodb import _ArangoDBBatchWriter
 from biocypher.output.write.graph._neo4j import _Neo4jBatchWriter
 from biocypher.output.write.graph._networkx import _NetworkXWriter
@@ -50,6 +51,7 @@ DBMS_TO_CLASS = {
     "Tabular": _PandasCSVWriter,
     "networkx": _NetworkXWriter,
     "NetworkX": _NetworkXWriter,
+    "airr": _AirrWriter,
 }
 
 

biocypher/output/write/graph/_airr.py

@@ -0,0 +1,32 @@
+"""Module to provide the AnnData writer class for BioCypher."""
+
+from biocypher._logger import logger
+from biocypher.output.write._writer import _Writer
+
+
+class _AirrWriter(_Writer):
+    """A minimal placeholder writer class that implements the required methods
+    but performs no actual writing operations, since there is an existing anndata native writer functionality
+    """
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        logger.info("Placeholder writer initialized")
+
+    def _write_node_data(self, nodes) -> bool:
+        """Required implementation that does nothing with nodes."""
+        logger.info("Placeholder: Node data received but not processed")
+        return True
+
+    def _write_edge_data(self, edges) -> bool:
+        """Required implementation that does nothing with edges."""
+        logger.info("Placeholder: Edge data received but not processed")
+        return True
+
+    def _construct_import_call(self) -> str:
+        """Return a placeholder import script."""
+        return "# This is a placeholder import script\nprint('No actual import functionality implemented')"
+
+    def _get_import_script_name(self) -> str:
+        """Return a placeholder script name."""
+        return "placeholder_import.py"

{biocypher-0.9.7.dist-info → biocypher-0.10.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: biocypher
-Version: 0.9.7
+Version: 0.10.1
 Summary: A unifying framework for biomedical research knowledge graphs
 Home-page: https://github.com/biocypher/biocypher
 License: MIT
@@ -19,6 +19,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
+Provides-Extra: scirpy
 Requires-Dist: PyYAML (>=5.0)
 Requires-Dist: appdirs
 Requires-Dist: more_itertools
@@ -27,6 +28,7 @@ Requires-Dist: networkx (>=3.0,<4.0)
 Requires-Dist: pandas (>=2.0.1,<3.0.0)
 Requires-Dist: pooch (>=1.7.0,<2.0.0)
 Requires-Dist: rdflib (>=6.2.0,<7.0.0)
+Requires-Dist: scirpy (>=0.22.0,<0.23.0) ; extra == "scirpy"
 Requires-Dist: tqdm (>=4.65.0,<5.0.0)
 Requires-Dist: treelib (==1.6.4)
 Project-URL: Bug Tracker, https://github.com/biocypher/biocypher/issues
@@ -79,7 +81,7 @@ Board](https://github.com/orgs/biocypher/projects/3/views/2).
 
 Install the package from PyPI using `pip install biocypher`. More comprehensive
 installation and configuration instructions can be found
-[here](https://biocypher.org/installation.html).
+[here](https://biocypher.org/BioCypher/installation/).
 
 Exemplary usage of BioCypher to build a graph database is shown in our tutorial
 and the various pipelines we have created. You can find these on the [Components

{biocypher-0.9.7.dist-info → biocypher-0.10.1.dist-info}/RECORD

@@ -11,7 +11,7 @@ biocypher/_deduplicate.py,sha256=rtglcaLRaVzNjLtaPwTGP8VvCM4PHYQ5CZ-cm32CrKQ,484
 biocypher/_get.py,sha256=_wUjhRjH2J6Qhq0Ndy3kdfaWhHDTT-dxyCvtuH36My4,14868
 biocypher/_logger.py,sha256=y9dh3SPJOCWXnkFSYSK7aj_-pB7zlAkNCf43Dp1lt74,2941
 biocypher/_mapping.py,sha256=ntspG2C_NaQODhWTBFk0CDvolkOCjtqlQ9E-NkJAuTg,9030
-biocypher/_metadata.py,sha256=MzfAA6RQw7SDP9sRRrQQbAUKSBX_VIpqeh10qERhTg4,1415
+biocypher/_metadata.py,sha256=dz_VeWCq0wx6Cr1hCzscvEVPLSRY1F8Tecu2saeAhkg,1416
 biocypher/_misc.py,sha256=YzlY7zwa0mim9QFg9HwXErkJFIH3cvLrbgjF8tKOIT8,6353
 biocypher/_ontology.py,sha256=lipZxU3aj6zrTbBrJZmCW6IRCuz-KQG3AfbYCVq6aFE,33133
 biocypher/_translate.py,sha256=NKSM9lxNjNNbgQrK_24eWYh3B41TS7kjnSwjySnK3s0,16851
@@ -20,15 +20,17 @@ biocypher/output/connect/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZ
 biocypher/output/connect/_get_connector.py,sha256=Qimv3kTkXYkhJZRT6nq8mwIM2wORCnyqqHqF2IByuuc,1152
 biocypher/output/connect/_neo4j_driver.py,sha256=kXjOXW12wZFfEp7plAuo40bPSvOfd-i9m4YaXoMq-p0,12357
 biocypher/output/in_memory/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-biocypher/output/in_memory/_get_in_memory_kg.py,sha256=29DvmOdRUuzEblyMxy5001R4zjEXW3eHNv0htko7c4Y,1115
-biocypher/output/in_memory/_in_memory_kg.py,sha256=g1TPN8PkeAyXbrRuTAjshqC8voI6EmLqR8S_otmviwU,1423
+biocypher/output/in_memory/_airr.py,sha256=wOYg2x1WCs7DTImtdxNyyoCmjDUVg1Ovh29kzMOQ688,20671
+biocypher/output/in_memory/_get_in_memory_kg.py,sha256=NQ-AT8jgAvb-aTK7zsYrItvlDEow_Mfni5tKEJjwjx0,1256
+biocypher/output/in_memory/_in_memory_kg.py,sha256=BTBtqb2ZC_zxXdOJ59BQKUoGXZSpiaRG6VecjZGWCm0,3526
 biocypher/output/in_memory/_networkx.py,sha256=cSOSAreP7S3oeGT6noZ1kAIvSnkVnU3NUp1OY4yqzn0,1515
-biocypher/output/in_memory/_pandas.py,sha256=Ot2jbK5t_YLHqw0BUv9Z_qWNy9r6IX1LYEyejOSJzos,3288
+biocypher/output/in_memory/_pandas.py,sha256=ndZcAAdsw38qZW3nWehcSxhpBGM8pXsn3DPoCcppI0U,1196
 biocypher/output/write/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 biocypher/output/write/_batch_writer.py,sha256=_Dao7z4KN0Uhr86oOOWYEDrIUikR7T0v1SJC2Btd8Y4,38745
-biocypher/output/write/_get_writer.py,sha256=JozRWCMhvh65aQAlcGiiD5x3Nl1HSW8mK1Zf2nTSOzI,4385
+biocypher/output/write/_get_writer.py,sha256=fXputhpt6K4mF8Ti6LFgwZIMU0GrK3aHkWIj1g4liwI,4469
 biocypher/output/write/_writer.py,sha256=y0dWI-RyQdrBLr9Fs91Y9KcCMjnlCaKJT0eWsIS2hG4,7158
 biocypher/output/write/graph/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+biocypher/output/write/graph/_airr.py,sha256=XWzcqwoMZKZ8n6f2Y77hv1ZSRyg2tOVqlzX133-biG8,1274
 biocypher/output/write/graph/_arangodb.py,sha256=xue3hm_DVB5pMR5qqfGXlXll3RpILA0tXos2J-as1-E,7906
 biocypher/output/write/graph/_neo4j.py,sha256=tBPhxn8JAmSS6KmiePofwr9LpGjHQH9BTnpHVK2ellM,12042
 biocypher/output/write/graph/_networkx.py,sha256=2WYkw5ZM3Bp236iwAxEAp3A1DxHKT4_hEPNMUKvPHp4,2320
@@ -38,7 +40,7 @@ biocypher/output/write/relational/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeR
 biocypher/output/write/relational/_csv.py,sha256=m0BSQXts88Qu5AEvoIgnwRz54ia38g4VN3PaA3LCYM8,2807
 biocypher/output/write/relational/_postgresql.py,sha256=75iJvv-0ewSsQXhVeoYoGnmYQnKY_B4iItZV7DpEBto,12190
 biocypher/output/write/relational/_sqlite.py,sha256=BuGWOeeNA83lbUvjpkzqcR9_baWLsbfmLXBKe4O1EPE,2105
-biocypher-0.9.7.dist-info/LICENSE,sha256=oejgxuxyjSnyPw3YPloz6-dCBB_nYizJ4jDQnr-xZUU,1082
-biocypher-0.9.7.dist-info/METADATA,sha256=uJ0y5nC0-_fSDjV9WTfVvMzGU7K0LunAkSmZrID1IYo,10621
-biocypher-0.9.7.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
-biocypher-0.9.7.dist-info/RECORD,,
+biocypher-0.10.1.dist-info/LICENSE,sha256=oejgxuxyjSnyPw3YPloz6-dCBB_nYizJ4jDQnr-xZUU,1082
+biocypher-0.10.1.dist-info/METADATA,sha256=Pxq7NgLmZcVn5zrge_QpB0xozDZWrtghWm2RMuOWZGU,10712
+biocypher-0.10.1.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
+biocypher-0.10.1.dist-info/RECORD,,