BioBridge-PrimeKG Multimodal Data Dump

In this tutorial, we will prepare Milvus database for storing and searching nodes and edges of a graph.

In particular, we are using PrimeKG multimodal data from the BioBridge project.

In [1]:

# Load necessary libraries
import os
import glob
import hydra
import cudf
import cupy as cp
from pymilvus import (
    db,
    connections,
    FieldSchema,
    CollectionSchema,
    DataType,
    Collection,
    utility,
    MilvusClient
)
from tqdm import tqdm
import time

# Load necessary libraries import os import glob import hydra import cudf import cupy as cp from pymilvus import ( db, connections, FieldSchema, CollectionSchema, DataType, Collection, utility, MilvusClient ) from tqdm import tqdm import time

Loading BioBridge-PrimeKG Multimodal Data¶

First, we need to get the path to the directory containing the parquet files of nodes and edges.

For nodes and edges, we have a separate folder that contains its enrichment and embeddings.

In [2]:

# Load hydra configuration
with hydra.initialize(version_base=None, config_path="../../../aiagents4pharma/talk2knowledgegraphs/configs"):
    cfg = hydra.compose(
        config_name="config", overrides=["tools/multimodal_subgraph_extraction=default"]
    )
    cfg = cfg.tools.multimodal_subgraph_extraction
cfg

# Load hydra configuration with hydra.initialize(version_base=None, config_path="../../../aiagents4pharma/talk2knowledgegraphs/configs"): cfg = hydra.compose( config_name="config", overrides=["tools/multimodal_subgraph_extraction=default"] ) cfg = cfg.tools.multimodal_subgraph_extraction cfg

Out[2]:

{'_target_': 'talk2knowledgegraphs.tools.multimodal_subgraph_extraction', 'ollama_embeddings': ['nomic-embed-text'], 'temperature': 0.1, 'streaming': False, 'topk': 5, 'topk_e': 5, 'cost_e': 0.5, 'c_const': 0.01, 'root': -1, 'num_clusters': 1, 'pruning': 'gw', 'verbosity_level': 0, 'node_id_column': 'node_id', 'node_attr_column': 'node_attr', 'edge_src_column': 'edge_src', 'edge_attr_column': 'edge_attr', 'edge_dst_column': 'edge_dst', 'node_colors_dict': {'gene/protein': '#6a79f7', 'molecular_function': '#82cafc', 'cellular_component': '#3f9b0b', 'biological_process': '#c5c9c7', 'drug': '#c4a661', 'disease': '#80013f'}, 'biobridge': {'source': '/mnt/blockstorage/biobridge_multimodal/', 'node_type': ['gene/protein', 'molecular_function', 'cellular_component', 'biological_process', 'drug', 'disease']}}

In [3]:

# You can set the source directory for biobridge data here
cfg.biobridge.source = "/mnt/blockstorage/biobridge_multimodal"

# You can set the source directory for biobridge data here cfg.biobridge.source = "/mnt/blockstorage/biobridge_multimodal"

In [4]:

%%time

# Loop over nodes and edges
graph_dict = {}
for element in ["nodes", "edges"]:
    # Make an empty dictionary for each folder
    graph_dict[element] = {}
    for stage in ["enrichment", "embedding"]:
        print(element, stage)
        # Create the file pattern for the current subfolder
        file_list = glob.glob(os.path.join(cfg.biobridge.source,
                                           element,
                                           stage, '*.parquet.gzip'))
        print(file_list)
        # Read and concatenate all dataframes in the folder
        # Except the edges embedding, which is too large to read in one go
        # We are using a chunk size to read the edges embedding in smaller parts instead
        if element == "edges" and stage == "embedding":
            # For edges embedding, only read two columns: triplet_index and edge_emb
            # graph_dict[element][stage] = cudf.concat([cudf.read_parquet(f, columns=["triplet_index", "edge_emb"]) for f in file_list[:2]], ignore_index=True)
            # Loop by chunks
            # file_list = file_list[:2]
            chunk_size = 5
            graph_dict[element][stage] = []
            for i in range(0, len(file_list), chunk_size):
                chunk_files = file_list[i:i+chunk_size]
                chunk_df = cudf.concat([cudf.read_parquet(f, columns=["triplet_index", "edge_emb"]) for f in chunk_files], ignore_index=True)
                graph_dict[element][stage].append(chunk_df)
        else:
            # For nodes and edges enrichment, read and concatenate all dataframes in the folder
            # This includes the nodes embedding, which is small enough to read in one go
            graph_dict[element][stage] = cudf.concat([cudf.read_parquet(f) for f in file_list], ignore_index=True)

%%time # Loop over nodes and edges graph_dict = {} for element in ["nodes", "edges"]: # Make an empty dictionary for each folder graph_dict[element] = {} for stage in ["enrichment", "embedding"]: print(element, stage) # Create the file pattern for the current subfolder file_list = glob.glob(os.path.join(cfg.biobridge.source, element, stage, '*.parquet.gzip')) print(file_list) # Read and concatenate all dataframes in the folder # Except the edges embedding, which is too large to read in one go # We are using a chunk size to read the edges embedding in smaller parts instead if element == "edges" and stage == "embedding": # For edges embedding, only read two columns: triplet_index and edge_emb # graph_dict[element][stage] = cudf.concat([cudf.read_parquet(f, columns=["triplet_index", "edge_emb"]) for f in file_list[:2]], ignore_index=True) # Loop by chunks # file_list = file_list[:2] chunk_size = 5 graph_dict[element][stage] = [] for i in range(0, len(file_list), chunk_size): chunk_files = file_list[i:i+chunk_size] chunk_df = cudf.concat([cudf.read_parquet(f, columns=["triplet_index", "edge_emb"]) for f in chunk_files], ignore_index=True) graph_dict[element][stage].append(chunk_df) else: # For nodes and edges enrichment, read and concatenate all dataframes in the folder # This includes the nodes embedding, which is small enough to read in one go graph_dict[element][stage] = cudf.concat([cudf.read_parquet(f) for f in file_list], ignore_index=True)

nodes enrichment
['/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/drug.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/cellular_component.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/gene_protein.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/disease.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/biological_process.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/enrichment/molecular_function.parquet.gzip']
nodes embedding
['/mnt/blockstorage/biobridge_multimodal/nodes/embedding/drug.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/embedding/cellular_component.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/embedding/gene_protein.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/embedding/disease.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/embedding/biological_process.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/nodes/embedding/molecular_function.parquet.gzip']
edges enrichment
['/mnt/blockstorage/biobridge_multimodal/edges/enrichment/edges.parquet.gzip']
edges embedding
['/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3050000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2250000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1150000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_600000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1550000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1750000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3100000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1600000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2750000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2950000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3700000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_500000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_800000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1050000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_300000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_250000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3600000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_900000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2400000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1100000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3450000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_400000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1500000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_0.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1250000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3900000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1400000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_50000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1450000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1800000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1000000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1950000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2100000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2350000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3500000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3750000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2800000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1350000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_150000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2500000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_450000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_850000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1850000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3000000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_550000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_200000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3300000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2650000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3350000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_100000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3400000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3550000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_350000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3800000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3200000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2850000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2200000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3850000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3250000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2900000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_750000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1300000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2300000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2700000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3150000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1900000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2150000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_950000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2450000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1700000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_650000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_3650000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2550000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_700000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2000000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2600000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_2050000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1200000.parquet.gzip', '/mnt/blockstorage/biobridge_multimodal/edges/embedding/edges_1650000.parquet.gzip']
CPU times: user 19.9 s, sys: 15.4 s, total: 35.3 s
Wall time: 24.4 s

In [5]:

# Get nodes enrichment and embedding dataframes
nodes_enrichment_df = graph_dict['nodes']['enrichment']
nodes_embedding_df = graph_dict['nodes']['embedding']

# Get edges enrichment and embedding dataframes
edges_enrichment_df = graph_dict['edges']['enrichment']
edges_embedding_df = graph_dict['edges']['embedding'] # !!consisted of a list of dataframes!!

# Get nodes enrichment and embedding dataframes nodes_enrichment_df = graph_dict['nodes']['enrichment'] nodes_embedding_df = graph_dict['nodes']['embedding'] # Get edges enrichment and embedding dataframes edges_enrichment_df = graph_dict['edges']['enrichment'] edges_embedding_df = graph_dict['edges']['embedding'] # !!consisted of a list of dataframes!!

In [6]:

# Merge nodes enrichment and embedding dataframes
merged_nodes_df = nodes_enrichment_df.merge(
    nodes_embedding_df[["node_id", "desc_emb", "feat_emb"]],
    on="node_id",
    how="left"
)
# del nodes_enrichment_df, nodes_embedding_df  # Free memory

# Merge nodes enrichment and embedding dataframes merged_nodes_df = nodes_enrichment_df.merge( nodes_embedding_df[["node_id", "desc_emb", "feat_emb"]], on="node_id", how="left" ) # del nodes_enrichment_df, nodes_embedding_df # Free memory

Normalization of the embeddings¶

Helper to normalize the vector/matrix.

In [2]:

def normalize_matrix(m, axis=1):
    """
    Normalize each row of a 2D matrix using CuPy.
    
    Parameters:
    m (cupy.ndarray): 2D matrix to normalize.
    
    Returns:
    cupy.ndarray: Normalized matrix.
    """
    norms = cp.linalg.norm(m, axis=axis, keepdims=True)
    return m / norms

def normalize_vector(v):
    """
    Normalize a vector using CuPy.
    
    Parameters:
    v (cupy.ndarray): Vector to normalize.
    
    Returns:
    cupy.ndarray: Normalized vector.
    """
    v = cp.asarray(v)
    norm = cp.linalg.norm(v)
    return v / norm

def normalize_matrix(m, axis=1): """ Normalize each row of a 2D matrix using CuPy. Parameters: m (cupy.ndarray): 2D matrix to normalize. Returns: cupy.ndarray: Normalized matrix. """ norms = cp.linalg.norm(m, axis=axis, keepdims=True) return m / norms def normalize_vector(v): """ Normalize a vector using CuPy. Parameters: v (cupy.ndarray): Vector to normalize. Returns: cupy.ndarray: Normalized vector. """ v = cp.asarray(v) norm = cp.linalg.norm(v) return v / norm

Setup Milvus Database¶

In [3]:

# Configuration for Milvus
milvus_host = "localhost"
milvus_port = "19530"
milvus_uri = "http://localhost:19530"
milvus_token = "root:Milvus"
milvus_user = "root"
milvus_password = "Milvus"
milvus_database = "t2kg_primekg"

# Configuration for Milvus milvus_host = "localhost" milvus_port = "19530" milvus_uri = "http://localhost:19530" milvus_token = "root:Milvus" milvus_user = "root" milvus_password = "Milvus" milvus_database = "t2kg_primekg"

In [4]:

# Connect to Milvus
connections.connect(
    alias="default",
    host=milvus_host,
    port=milvus_port,
    user=milvus_user,
    password=milvus_password
)

# Connect to Milvus connections.connect( alias="default", host=milvus_host, port=milvus_port, user=milvus_user, password=milvus_password )

In [5]:

# Check if the database exists, create if it doesn't
if milvus_database not in db.list_database():
    db.create_database(milvus_database)

# Switch to the desired database
db.using_database(milvus_database)

# Check if the database exists, create if it doesn't if milvus_database not in db.list_database(): db.create_database(milvus_database) # Switch to the desired database db.using_database(milvus_database)

In [11]:

# List all collections
for coll in utility.list_collections():
    print(f"Collection: {coll}")

    # Load the collection to get stats
    collection = Collection(name=coll)
    print(collection.num_entities)

# List all collections for coll in utility.list_collections(): print(f"Collection: {coll}") # Load the collection to get stats collection = Collection(name=coll) print(collection.num_entities)

In [12]:

# A helper function to chunk the data into smaller parts
# Utility: chunk generator
def chunked(data_list, chunk_size):
    for i in range(0, len(data_list), chunk_size):
        yield data_list[i:i + chunk_size]

# A helper function to chunk the data into smaller parts # Utility: chunk generator def chunked(data_list, chunk_size): for i in range(0, len(data_list), chunk_size): yield data_list[i:i + chunk_size]

Building Node Collection (Description Embedding)¶

In [13]:

%%time

# Configuration for Milvus collection
node_coll_name = f"{milvus_database}_nodes"

# Define schema for the collection
# Leave out the feat and feat_emb fields for now
desc_emb_dim = len(merged_nodes_df.iloc[0]['desc_emb'].to_arrow().to_pylist()[0])
node_fields = [
    FieldSchema(name="node_index", dtype=DataType.INT64, is_primary=True),
    FieldSchema(name="node_id", dtype=DataType.VARCHAR, max_length=1024),
    FieldSchema(name="node_name", dtype=DataType.VARCHAR, max_length=1024,
                enable_analyzer=True, enable_match=True),
    FieldSchema(name="node_type", dtype=DataType.VARCHAR, max_length=1024,
                enable_analyzer=True, enable_match=True),
    FieldSchema(name="desc", dtype=DataType.VARCHAR, max_length=40960,
                enable_analyzer=True, enable_match=True),
    FieldSchema(name="desc_emb", dtype=DataType.FLOAT_VECTOR, dim=desc_emb_dim),
]
schema  = CollectionSchema(fields=node_fields, description=f"Schema for collection {node_coll_name}")

# Create collection if it doesn't exist
if not utility.has_collection(node_coll_name):
    collection = Collection(name=node_coll_name, schema=schema)
else:
    collection = Collection(name=node_coll_name)

# Create indexes
collection.create_index(
    field_name="node_index",
    index_params={"index_type": "STL_SORT"}, # STL_SORT
    index_name="node_index_index"
)
# Create index for node_name, node_type, desc fields (inverted)
collection.create_index(
    field_name="node_name",
    index_params={"index_type": "INVERTED"},
    index_name="node_name_index"
)
collection.create_index(
    field_name="node_type",
    index_params={"index_type": "INVERTED"},
    index_name="node_type_index"
)
collection.create_index(
    field_name="desc",
    index_params={"index_type": "INVERTED"},
    index_name="desc_index"
)
collection.create_index(
    field_name="desc_emb",
    index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX
    index_name="desc_emb_index"
)

# Prepare data for insertion
# Normalize the embeddings
graph_desc_emb_cp = cp.asarray(merged_nodes_df["desc_emb"].list.leaves).astype(cp.float32).reshape(merged_nodes_df.shape[0], -1)
graph_desc_emb_norm = normalize_matrix(graph_desc_emb_cp, axis=1)
data = [
    merged_nodes_df["node_index"].to_arrow().to_pylist(),
    merged_nodes_df["node_id"].to_arrow().to_pylist(),
    merged_nodes_df["node_name"].to_arrow().to_pylist(),
    merged_nodes_df["node_type"].to_arrow().to_pylist(),
    merged_nodes_df["desc"].to_arrow().to_pylist(),
    graph_desc_emb_norm.tolist(), # Use normalized embeddings
]

# Insert data in batches
batch_size = 500
total = len(data[0])
for i in tqdm(range(0, total, batch_size)):
    batch = [
        col[i:i+batch_size] for col in data
    ]
    collection.insert(batch)

# Flush to persist data
collection.flush()

# Get collection stats
print(collection.num_entities)

%%time # Configuration for Milvus collection node_coll_name = f"{milvus_database}_nodes" # Define schema for the collection # Leave out the feat and feat_emb fields for now desc_emb_dim = len(merged_nodes_df.iloc[0]['desc_emb'].to_arrow().to_pylist()[0]) node_fields = [ FieldSchema(name="node_index", dtype=DataType.INT64, is_primary=True), FieldSchema(name="node_id", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="node_name", dtype=DataType.VARCHAR, max_length=1024, enable_analyzer=True, enable_match=True), FieldSchema(name="node_type", dtype=DataType.VARCHAR, max_length=1024, enable_analyzer=True, enable_match=True), FieldSchema(name="desc", dtype=DataType.VARCHAR, max_length=40960, enable_analyzer=True, enable_match=True), FieldSchema(name="desc_emb", dtype=DataType.FLOAT_VECTOR, dim=desc_emb_dim), ] schema = CollectionSchema(fields=node_fields, description=f"Schema for collection {node_coll_name}") # Create collection if it doesn't exist if not utility.has_collection(node_coll_name): collection = Collection(name=node_coll_name, schema=schema) else: collection = Collection(name=node_coll_name) # Create indexes collection.create_index( field_name="node_index", index_params={"index_type": "STL_SORT"}, # STL_SORT index_name="node_index_index" ) # Create index for node_name, node_type, desc fields (inverted) collection.create_index( field_name="node_name", index_params={"index_type": "INVERTED"}, index_name="node_name_index" ) collection.create_index( field_name="node_type", index_params={"index_type": "INVERTED"}, index_name="node_type_index" ) collection.create_index( field_name="desc", index_params={"index_type": "INVERTED"}, index_name="desc_index" ) collection.create_index( field_name="desc_emb", index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX index_name="desc_emb_index" ) # Prepare data for insertion # Normalize the embeddings graph_desc_emb_cp = cp.asarray(merged_nodes_df["desc_emb"].list.leaves).astype(cp.float32).reshape(merged_nodes_df.shape[0], -1) graph_desc_emb_norm = normalize_matrix(graph_desc_emb_cp, axis=1) data = [ merged_nodes_df["node_index"].to_arrow().to_pylist(), merged_nodes_df["node_id"].to_arrow().to_pylist(), merged_nodes_df["node_name"].to_arrow().to_pylist(), merged_nodes_df["node_type"].to_arrow().to_pylist(), merged_nodes_df["desc"].to_arrow().to_pylist(), graph_desc_emb_norm.tolist(), # Use normalized embeddings ] # Insert data in batches batch_size = 500 total = len(data[0]) for i in tqdm(range(0, total, batch_size)): batch = [ col[i:i+batch_size] for col in data ] collection.insert(batch) # Flush to persist data collection.flush() # Get collection stats print(collection.num_entities)

100%|██████████| 170/170 [01:52<00:00,  1.51it/s]

84981
CPU times: user 28.2 s, sys: 3.68 s, total: 31.9 s
Wall time: 2min 3s

In [14]:

# List all collections
for coll in utility.list_collections():
    print(f"Collection: {coll}")

    # Load the collection to get stats
    collection = Collection(name=coll)
    print(collection.num_entities)

# List all collections for coll in utility.list_collections(): print(f"Collection: {coll}") # Load the collection to get stats collection = Collection(name=coll) print(collection.num_entities)

Collection: t2kg_primekg_nodes
84981

In [9]:

%%time

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_nodes')

# Load the collection into memory before query
collection.load()

# Query by expr on node_index
expr = "node_index == 13814"
output_fields = ["node_index", "node_id", "node_name", "node_type", "desc", "desc_emb"]

results = collection.query(expr, output_fields=output_fields)

print(results)

%%time # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_nodes') # Load the collection into memory before query collection.load() # Query by expr on node_index expr = "node_index == 13814" output_fields = ["node_index", "node_id", "node_name", "node_type", "desc", "desc_emb"] results = collection.query(expr, output_fields=output_fields) print(results)

data: ["{'node_name': 'Copper', 'node_type': 'drug', 'desc': 'Copper belongs to drug node. Copper is a transition metal and a trace element in the body. It is important to the function of many enzymes including cytochrome c oxidase, monoamine oxidase and superoxide dismutase. Copper is commonly used in contraceptive intrauterine devices (IUD). Copper is absorbed from the gut via high affinity copper uptake protein and likely through low affinity copper uptake protein and natural resistance-associated macrophage protein-2. It is believed that copper is reduced to the Cu1+ form prior to transport. Once inside the enterocyte, it is bound to copper transport protein ATOX1 which shuttles the ion to copper transporting ATPase-1 on the golgi membrane which take up copper into the golgi apparatus. Once copper has been secreted by enterocytes into the systemic circulation it remain largely bound by ceruloplasmin (65-90%), albumin (18%), and alpha 2-macroglobulin (12%).  Copper is nearly entirely bound by ceruloplasmin (65-90%), plasma albumin (18%), and alpha 2-macroglobulin (12%). Copper is incorporated into many enzymes throughout the body as an essential part of their function. Copper ions are known to reduce fertility when released from copper-containing IUDs. For use in the supplementation of total parenteral nutrition and in contraception with intrauterine devices.', 'desc_emb': [-0.008437114, -0.006676033, 0.017289387, -0.031070005, -0.02470869, 0.026931802, -0.025539009, -0.016030516, -0.022900736, -0.017262602, 0.01635193, 0.02225791, -0.013927933, 0.0072853803, -0.010954855, 0.01364, 0.033105627, 0.025713107, -0.0011316446, 0.019137517, -0.042480197, 0.02196328, 0.0012337605, -0.011530722, 0.001766102, 0.0112494845, 0.030427177, -0.03964104, -0.0038636622, -0.02178918, -0.00078553875, 0.0020724498, -0.02006158, 0.01814649, 0.010546392, 0.011095474, 0.019512499, 0.006207305, -0.002527786, -0.0065588513, 0.018280413, 0.006548807, -0.0063512716, 0.024239961, -0.0007826092, 0.009568758, 0.0057821013, 0.0011140673, -0.025391694, 0.0064517134, 0.009200471, 0.012454786, -0.0059997253, -0.015856417, -0.0007487101, -0.021869535, 0.0077942857, 0.018454513, 0.011999451, -0.02035621, -0.0113432305, -0.010820933, -0.011624468, 0.024373883, 0.00061897276, 0.03139142, 0.0070309276, 0.008109003, -0.0015032793, 0.017235817, 0.0064718015, 0.031150358, 0.012809681, -0.012829769, 0.037096515, -0.025820246, -0.065086305, 0.00038105118, 0.003024973, 0.01802596, 0.019057162, 0.0065119783, -0.0011835395, 0.01903038, 0.01008436, 0.015923379, 0.008470594, 0.025418477, -0.004456269, -0.0039708004, -0.0071246736, 0.01754384, 0.019887483, 0.03682867, -0.006200609, 0.03460556, -0.016418891, 0.050461974, -0.013064133, -0.011738301, -0.00844381, 0.01552161, -0.009287521, -0.01676709, -0.019485714, 0.005273196, -0.0142895235, -0.011678036, 0.016981365, -0.022606106, -0.011236092, 0.071407445, 0.010526303, -0.0386768, -0.008309888, 0.0015719145, 0.027427314, -0.007051016, -0.00069723366, -0.0017995827, -0.016539421, 0.014959136, 0.01676709, -0.023329288, 0.009401355, 0.011992754, -0.022351654, -0.019244654, 0.007546529, -0.007821071, 0.0013810751, -0.0016715194, 0.003853618, 0.04004281, -0.013372155, 0.0042955624, -0.020275857, -0.0016430608, -0.005882543, -0.024119431, 0.03361453, 0.021280276, -0.011952577, -0.0022632892, 0.0011249485, 0.040551715, 0.038891077, 0.008972803, -0.009321001, 0.004888169, 0.009890172, -0.0013693569, 0.030534314, -0.028739754, 0.021574905, 0.0130507415, -0.0073791263, 0.016218007, 0.014061855, -0.029355798, -0.0016413869, -0.035221603, -0.009990614, -0.0026014433, 0.017342957, 0.01982052, 0.012829769, -0.0066392045, 0.025726499, -0.005929416, 0.01220703, 0.02304805, -0.029543288, 0.031284284, 0.00592272, 0.009622327, 0.007881336, 0.014222562, -0.013767227, -0.029329013, -0.0033681493, -0.0037900049, 0.018722357, -0.009321001, -0.009856692, 0.0010152995, 0.018333983, 0.012896731, 0.011557506, -0.015856417, -0.027454099, 0.008450506, -0.029302228, -0.027748728, -0.63554233, 0.0041382033, -0.003960756, -0.021936495, -0.01101512, 0.019606246, 0.017704546, -0.004278822, 0.008852273, 5.335972e-05, 0.000274541, 0.023623917, 0.020757977, -0.023155188, -0.012046323, -0.02494975, -0.015267157, -0.0025411781, 0.006274266, -0.053033285, -0.052872576, 0.0046169762, -0.020570487, 0.016981365, 0.014329701, -0.002574659, 0.018575042, -0.019726776, -0.009347786, 0.0075063524, 0.002807349, 0.023356073, 0.019258047, 0.042399842, 0.051988687, 0.0048781247, -0.014878782, 0.020275857, -0.0076871477, 0.03503411, -0.010037486, -0.024963142, 0.015789455, -0.040016025, -0.0047944235, 0.011537418, 0.019539284, 0.018842887, 0.019673206, -0.006160432, -0.012963692, -0.015280549, -0.011959273, 0.0075800098, -0.00516271, 0.011617771, 0.029114736, -0.026034521, 0.008102308, -0.004181728, 0.001487376, -0.0014020005, 0.00071020745, -0.01766437, -0.019766951, -0.010995032, -0.036266197, -0.008731743, 0.0051392736, 0.028793324, 0.011075386, 0.012474875, 0.009749553, -0.016325146, 0.030507531, 0.015481433, -0.008169269, 0.007760805, -0.0039473637, 0.04620324, 0.013459205, 0.012059716, -0.008216142, -0.017744724, 0.030695021, 0.007519745, -0.006481846, -0.00019094409, 0.020369602, -0.018200058, 0.009454924, 0.0076201865, -0.0066392045, -0.0193384, -0.025391694, 0.018936632, -0.019231262, -0.022177555, 0.019807128, 0.0041482477, -0.0082094455, -0.027333569, -0.035194818, 0.015401079, 0.016151046, 0.014128816, -0.026851447, -0.0047475505, 0.038408954, -0.024333706, -0.0041348552, -0.0017008149, 0.0075264405, -0.016807266, -0.0034551988, -0.0041214633, 0.003189028, 0.012836466, -0.0002423159, -0.027855866, 0.030427177, -0.011510633, 0.0033396906, 0.007539833, -0.010057575, 0.0055443887, 0.0026717526, -0.02524438, -0.0032911438, -0.014008286, 0.002204698, -0.0062943543, 0.0224454, -0.010747275, 0.03179319, 0.013619912, 0.03969461, -0.0013752161, -0.0063613155, -0.03964104, -0.016700128, -0.011611075, 0.006870221, -4.9462367e-05, -0.012394521, -0.015079666, -0.023623917, 0.02453459, 0.003016603, -0.012481571, -0.017168857, -0.006826696, -0.0050220913, 0.03755185, 0.0193384, -0.012173549, -0.009200471, -0.0056950515, -0.01262219, -0.0032057683, -0.008470594, 0.038408954, -0.020329425, 0.005705096, -0.009374571, 0.0071849385, 0.0056749634, 0.022485577, -0.013593127, -0.03806076, 0.006421581, 0.0027654984, -0.013994894, -0.008504075, -0.025391694, 0.0005382008, -0.0013676828, -0.005279892, -0.02006158, 0.0006177172, -0.0077273245, -0.011323142, -0.014945744, -0.009622327, 0.022498969, 0.014557368, 0.01921787, 0.0050890525, 0.004061198, 0.027079117, 0.0032308786, 0.014463623, -0.021601688, -0.008852273, -0.010499519, -0.0053535495, -0.0070778006, 0.0051928423, -0.002079146, 0.02213738, 0.014222562, -0.012253903, -0.006615768, -0.034391284, 0.00048756137, -0.009957133, -0.0130172605, 0.005427207, 0.03002541, -0.0023955377, -0.005718488, -0.021333843, 0.0013534536, -0.013077525, -0.0031655915, 0.042239137, 0.0065387627, -0.018950025, -0.00015662647, -0.025083672, -0.0006281799, -0.010887894, 0.036266197, -0.0013852601, -0.0011827025, 0.0016397128, 0.02035621, -0.0025880511, -0.0035589887, -0.0057620127, 0.01101512, -0.009354482, -0.009287521, -0.010111144, -0.0024323664, 0.0072050267, 0.007345645, -0.011336534, 0.032409232, -0.015320727, -0.0017995827, 0.010532999, 0.028257633, -0.00056498527, 0.008745135, -0.013820795, 0.020811547, 0.018333983, 0.0041382033, 0.010298636, 0.0005545226, 0.011303053, -0.03790005, -0.016137654, 0.0016070692, -0.01020489, 0.012595405, 0.030105762, 0.035382308, 0.028043358, 0.030802159, -0.013251625, 0.0038435739, -0.013070829, 0.0035589887, 0.021052606, 0.017342957, -0.016378714, -0.019164301, 0.00043692195, -0.036667965, -0.00024566398, 0.009394659, -0.019740168, 0.013613216, 0.03144499, -0.01790543, 0.01724921, 0.013646696, -0.011202612, -0.028686184, -0.027266607, 0.0076268828, 0.019646421, -0.01032542, -0.026195228, -0.029114736, 0.015976947, 0.004928346, 0.01856165, -0.009508493, 0.019258047, -0.010466038, 0.025324732, -0.017624194, -0.0038804025, 0.021119568, -0.0087987045, 0.010921375, -0.012267295, -0.016619775, -0.021976672, -0.041676663, 0.0151600195, 0.053568974, 0.007713932, 0.011912401, -0.0077474127, 0.009762946, 0.01110217, -0.016392106, 0.0046872855, 0.0025227638, -0.04087313, 0.00070727785, 0.01821345, -0.020048188, 0.014075248, 0.030962866, -0.022003457, -0.018347373, -0.0094817085, -0.0050220913, 0.001211998, 0.012736023, 0.020034797, -0.01343242, 0.0070577124, 0.019258047, 0.0061738244, 0.0067563867, -0.01581624, 0.019311616, 0.0057352283, 0.010452646, -0.021467766, -0.012140069, -0.0019602897, -0.0013844232, 0.0045500146, -0.00853086, -0.023141798, 0.0020172067, -0.0075733135, 0.021507943, -0.021601688, 0.011229396, 0.023650702, 0.013901149, -0.0015576853, 0.006702818, 0.02871297, 0.03701616, 0.00019502455, 0.024963142, -0.0034250661, 0.01802596, 0.010198194, -0.0448908, -0.0026667304, 0.009776338, 0.016057301, -0.0027956308, 0.022271302, 0.017101895, 0.026971979, 0.009454924, -0.009756249, -0.01032542, -0.020865116, 0.012675758, -0.01163786, 0.0040310654, -0.010211586, 0.018066136, -0.000767543, 0.011651251, -0.034525204, 0.028016573, -0.028793324, 0.011216004, 0.010419166, -0.018427728, -0.0053736377, -0.0022381789, -0.011557506, 0.02938258, -0.007901424, 0.019807128, -0.015119842, -0.012494964, -0.029007599, -0.0042587337, 0.0060231616, 0.011216004, -0.030828943, 0.00090062834, -0.010332116, 0.017048327, 0.02686484, -0.0006809119, -0.020624055, 0.0066860775, 0.03026647, -0.0037866568, -0.02208381, 0.0014647766, -0.03800719, 0.011242788, -0.008276407, 0.037444714, -5.63939e-05, -0.013727049, -0.0001843526, 0.008510771, -0.022539146, -0.009756249, -0.00853086, -0.0043123025, 0.008571036, 0.011416888, 0.010405773, 0.008423721, -0.012501659, -0.00015045349, -0.011115562, -0.0003496631, -0.03313241, -0.011055297, -0.005514256, -0.0120864995, 0.006100167, -0.015776064, -0.02620862, -0.00028395743, -0.0097294655, -0.009990614, -0.016914405, 0.00078302773, 0.009079942, -0.011115562, 0.03356096, -0.01196597, -0.010767364, -0.0055912617, 0.007821071, 0.025940776, 0.023128405, -0.009769642, 0.020878509, -0.009716073, -0.013820795, -0.0048781247, 0.0052397153, 0.020985646, 0.032007463, -0.020691017, -0.032650292, -0.0014061856, -0.0060399016, -0.0091402065, -0.015468041, -0.0065722433, -0.0004624509, -0.007258596, 0.009033068, -0.011751694, -0.012320864, 0.0021193225, -0.025230987, 0.006813304, 0.01474486, -0.0061302995, 0.00942814, 0.012722631, -0.027775513, -0.031177143, 0.01957946, -0.015655532, -0.018066136, -0.007091193, -0.022565931, 0.04012316, 0.025177417, 0.025806854, 0.0005172754, 0.0060131173, -0.005109141, -0.00176108, 0.016365321, -0.010820933, -0.00090397644, 0.0082697105, 0.015843024, 0.015173412, -0.010352205, 0.01420917, -0.0014145557, -0.008021954, 0.015548394, -0.017450094, -0.0012814703, -0.022820383, -0.0072518997, -0.04269447, -0.004000933, -0.02130706, -0.0032660333, -0.031284284, -0.01844112, 0.014811821, 0.008792008, 0.019766951, -0.019900873, 0.028311202, -0.02375784, 0.035489447, 0.007218419, 0.0032040942, -0.005293284, -0.026971979, -0.00540377, 0.014276131, 0.020637447, -0.002810697, 0.015293942, -0.0067965635, -0.03098965, 0.0056548747, 0.012977084, -0.014235955, -0.007432695, 0.020034797, -0.022887344, -0.0052664997, -0.0039841924, -0.01569571, -0.046953205, -0.00029965147, 0.023128405, -0.0059093274, 0.018293805, -0.008919234, -0.0006009769, 0.02208381, -0.006073382, 0.042426627, -0.021588298, 0.038730368, 0.0048111635, -0.0042955624, -0.0052865883, -0.015762672, 0.0068467846, -0.002420648, 0.021333843, 0.04004281, -0.0011885617, -0.003135459, -0.011082081, -0.0072452035, 0.0050086994, -0.014557368, 0.007941601, 0.0017393176, 0.005105793, -0.02817728, -0.000826971, 0.0025177419, 0.00901298, 0.00045073268, -0.012997172, -0.021815965, 0.009274129, -0.0037129994, 0.018106313, 0.0016405499, 0.02627558, 0.0027839127, -0.00083534117, -0.00031513625, -0.027909435, -0.011631164, -0.0077206283, 0.017557232, 0.024293529, 0.014423446, 0.013405636, -8.83051e-05, -0.0039339717, 0.011216004, -0.020074973, 0.002909465, 0.036159057, 0.0059528523, 0.013927933, 0.025030103, 0.009274129, 0.0033296465, -0.029007599, -0.0013224841, -0.0047944235, -0.024132824, -0.015240373, -0.0040879827, 0.00015348766, -0.029462935, -0.016981365, -0.0016397128, 0.0062374375, -0.049819145, 0.0063043986, 0.049229886, -0.040149946, -0.014302916, -0.0050455276, 0.001155918, 0.029570073, -0.0095419735, -0.009334394, -0.029409366, 0.016070694, -0.010245066, 0.030828943, 0.026730917, 0.024333706, -0.004007629, 0.004720766, -0.015668925, -0.018722357, 0.013131095, -0.0072050267, -0.030507531, 0.0042754738, -0.013124399, -0.0040277173, 0.035060894, -0.0087049585, 0.015494825, 0.0010822607, 0.013994894, 0.0012019539, -0.009796427, 0.007546529, -0.029543288, -0.0004406885, 0.018963417, -0.011959273, -0.026222013, -0.012588709, 0.0012873295, 0.0110017285, -0.018588435, -0.0029914924, -0.023998901, 0.0077273245, -0.021561513, -0.020691017, 0.014316308, -0.00514597, -0.018133098, 0.015146627, -0.005879195, 0.003413348, -0.023798017, -0.015856417, 0.01850808, -0.00082320446, -0.0021025822, 0.00571514, 0.0059762886, -0.02166865, -0.02823085, 0.019043772, 0.0062608737, 0.01873575, 0.000115717354, -0.026530035, -0.0047676386, 0.018226843, -0.020436564, -0.027092509, -0.0053669414, 0.024266746, -0.0014061856, 0.023195365, 0.0076737553, -0.01562875, -0.008865666, 0.025324732, -0.039078567, 0.0077407164, 0.03425736, 0.00061939127, 0.033694886, 0.013345371, -0.038998213, -0.024936358, 0.020945469, 0.00045575478, 0.004734158, 0.014142209, -0.0091402065, 3.59655e-05, -0.012133373, -0.018066136, 2.0389169e-05, -0.011912401, 0.03198068, -0.007559921, -0.018494688, 0.0070711044, 0.002855896, -0.0065555032, -0.0012019539, -0.007586706, 0.014222562, -0.013994894, 0.0025512225, -0.0067362986, 0.03664118, -0.009602238, -0.0117784785, 0.01599034, -0.0017142071, -0.001816323, 0.023382857, -0.017222425, -0.023423035, -0.011269573, -0.026489858, -0.00024524546, -0.008604516, 0.005926068, -0.008490683, 0.0024691948, -0.001923461, -0.02782908, 0.010178105, -0.018575042, -0.01469129, -0.028766539, 0.012977084, -0.008450506, 0.0042085126, -0.0014672877, -0.0024457586, -0.0030048846, -0.015601964, -0.004760943, 0.0046605007, -0.023007875, -0.025913991, -0.009187079, 0.0075733135, 0.01707511, -0.019499106, -0.006809956, 0.00847729, 0.017235817, 0.19231261, -0.0063881003, 0.028391555, 0.027320176, -0.009441532, 0.0026449682, 0.015401079, 0.0048145116, 0.020503525, 0.015950162, 0.013218144, 0.007432695, -0.014543977, 0.0008219489, -0.03382881, -0.015642142, -0.042426627, 0.0019619637, -0.012709239, 0.017958999, -0.013211448, -0.020449957, 0.023409642, -0.014651114, 0.008062131, -0.0011977688, 0.010720491, -0.0028960726, 0.019016987, 0.00032894698, 0.0017778203, 0.005390378, 0.0108477175, -0.00043273685, 0.014168994, -0.0067329505, 0.011450368, -0.028552262, 0.01224051, -0.019445539, -0.0039774966, 0.010633442, -0.028257633, -0.010834325, -0.01023837, 0.013265017, 0.00040888193, -0.023141798, -0.014677899, -0.0045801476, -0.035998352, 0.010961551, 0.0021544772, 0.0027219735, -0.0039339717, 0.012695847, -0.003026647, 0.0005666593, 0.008671478, 0.0038770544, -0.0035824252, -0.0019586156, -0.012508355, 0.040685635, -0.0028207414, 0.030373607, -0.023208758, -0.009488405, 0.02679788, -0.024976535, -0.0007658689, 0.008229534, -0.0071380655, 0.00011770526, -0.0041984683, 0.007425999, -0.01044595, 0.017932214, 0.0131110065, -0.00018236469, -0.010767364, -0.01110217, -0.029864702, 0.013680177, -0.006200609, -0.0011400147, 0.014651114, -0.0065320665, -0.025873814, -0.009441532, 0.012106588, -0.019847305, -0.00853086, -0.012715935, 0.013305194, 0.0055778692, 0.011416888, 0.0060131173, -0.002914487, 0.0052899364, -0.023838194, 0.072318114, 0.009086638, 0.014530584, 0.01023837, 0.01706172, -0.021119568, 0.01880271, 0.0057553165, -0.021923102, 0.017209033, -0.02662378, -0.0134525085, -0.011416888, 0.0065655475, 0.025646146, 0.018775925, -0.011905705, 0.010419166, -0.009146903, 0.0070108394, -0.024400668, 0.00925404, 0.004181728, -0.0052999803, 1.2934501e-05, -0.027373746, 0.015414472, -0.020543702, -0.035757292, -0.011631164, -0.008135788, 0.0059327637, 0.009930349, -0.012454786, 0.020798154, 0.01730278, -0.004847992, 0.004690633, 0.024065861, -0.01826702, -0.003512116, 0.028980814, 0.0059695924, 0.005343505, -0.020744586, -0.02154812, -0.00063445757, -0.02220434, -0.019793736, 0.001487376, -0.017731331, -0.0049417377, -0.0005821441, 0.013847579, -0.022632891, -0.009508493, -0.036185842, 0.01396811, -0.017570624, -0.016606383, 0.001157592, 0.028820107, 0.0021193225, -0.013847579, 0.009180383, -0.16874227, 0.015588571, 0.011082081, -0.027400529, 0.013425724, 0.0067496905, -0.02381141, 0.01844112, 0.0076871477, -0.011162435, 0.033587746, 0.008571036, -0.022177555, -0.022659676, -0.003512116, -0.008149181, -0.006542111, 0.005323417, 0.021333843, 0.01635193, 0.023061443, -0.027668376, 0.0059662443, -0.009829907, 0.03431093, 0.027146077, 0.017383132, -0.0064483653, 0.009006284, -0.009448228, -0.00042018163, 0.019191084, 0.0417838, 0.00017807499, 0.025043495, 0.004837948, 0.017275995, -0.026637172, 0.00018121379, 0.022780206, 0.0069103977, 0.0014488733, -0.014543977, 0.02698537, -0.020945469, 0.03187354, -0.012896731, 0.002855896, 0.029329013, 0.002906117, 0.026945194, -0.003800049, -0.0041448995, 0.024065861, 0.01999462, 0.0019000245, -0.0012990476, 0.036453687, -0.0031388071, -0.011122258, 0.0028358076, -0.017450094, 0.017195642, 0.0009968851, 0.023530172, -0.015079666, -0.0011492219, 0.019070555, -0.025391694, -0.002743736, -0.012756112, 0.008169269, 0.018789317, -0.010365597, 0.019793736, 0.019043772, -0.019766951, 0.0050823563, -0.003356431, 0.015950162, -0.0072719883, 0.009937045, -0.038757153, -0.012006146, -0.016365321, -0.01574928, 0.01683405, -0.0051359255, -0.004653805, 0.0034418066, 0.017329564, -0.02137402, -0.031311065, -0.03157891, 0.026596995, 0.01310431, 0.014343092, -0.009655808, -0.01844112, -0.022070417, 0.025592577, -0.021226706, -0.029811133, -0.0023938636, 0.044676524, 0.012494964, -0.01184544, -0.017048327, 0.02782908, -0.0025076976, 0.009849995, -0.0013316913, 0.0193384, 0.040149946, -0.020195503, 0.034926973, 0.020141935, -0.020824939, -0.010004006, -0.020262465, 0.045640767, 0.007499656, -0.00045240673, 0.010720491, -0.026288973, -0.0073992144, -0.11935167, -0.0065588513, 0.02513724, 0.013927933, -0.009361179, 0.0033095581, -0.01903038, 0.029275443, 0.013666784, 0.022485577, -0.024427451, -0.016726913, -0.009896868, -0.005882543, 0.008872362, 0.009207168, -0.009501797, -0.0030015365, -0.027775513, 0.032141387, -0.001972008, -0.023074836, 0.006398144, 0.01861522, -0.008109003, -0.010499519, -0.005038832, 0.0130172605, 0.0058658025, 0.0071447617, 0.02441406, 0.00042562225, 0.035194818, -0.0025076976, -0.016874228, 0.007004143, 0.018695572, 0.00035803326, 0.044542603, -0.02016872, -0.004563407, 0.010620049, 0.0044730096, -0.011510633, 0.026061306, -0.010432558, -0.037096515, -0.005440599, -0.014450231, -0.011765086, -0.025954168, -0.0095419735, -0.03736436, -0.0046337163, 0.011818655, 0.005946156, 0.021226706, 0.019191084, -0.029757565, -0.013171271, -0.0018447816, -0.0039071874, -0.012227118, 0.007452783, 0.031016435, -0.014356485, -0.008417025, -0.002807349, 0.00208919, -0.0405785, -0.025030103, -0.0056816596, -0.016606383, -0.0039205793, -0.019164301, 0.04028387, 0.019351792, 0.012890034, 0.01969999, -0.041703448, -0.011865527, -0.024373883, -0.0050723124, -0.033802025, -0.0056247422, 0.038542878, 0.0364269, 0.0030567795, 0.009742857, -0.019432146, -0.019351792, 0.036989376, 0.0048982133, -0.015334118, 0.0010881198, 0.043015886, 0.0024256702, -0.00086128863, 0.018226843, 0.007981777, -0.016485853, -0.024132824, -0.049712006, 0.0054707313, 0.025913991, -0.0067999116, 0.00084998895, 0.00028479443, 0.0081156995, 0.004071242, -0.0093879625, 0.012970388, -0.010365597, -0.004821208, 0.005705096, -0.011979362, -0.02494975, 0.0039272755, 0.033052057, -0.0040879827, 0.0024859353, 0.008182661, -0.014932351, 0.0040411097, -0.0035991655, 0.013445812, -0.042828396, -0.0010069293, 0.0075532254, -0.013767227, -0.0077808932, -0.006106863, 0.0077675013, -0.035596583, -0.0016296686, -0.0072719883, -0.005122533, -0.028445125, -0.01486539, -0.003245945, 0.014396662, 0.02161508, -0.00023792157, -0.023315895, -0.0058591063, -0.011651251, 0.004235297, -0.007024232, -0.026637172, 0.010064271, -0.006763083, 0.013044045, 0.026596995, 0.012126677, 0.014905566, -0.032462798, -0.0009801448, -0.0042520375, -0.003525508, -0.0138877565, -0.0343645, -0.02058388, 0.011564203, 0.011021816, -0.0034016299, 0.00616378, 0.029409366, -0.001814649, -0.027614806, -0.029275443, 0.030400392, -0.007218419, -0.011724909, -0.016445676, -0.004653805, 0.0107539715, 0.013660088, 0.0042520375, -0.021467766, 0.002519416, -0.004667197, 0.013144487, -0.017958999, -0.014597545, -0.013633303, 0.022726638, 0.004114767, 0.0128632495, 0.006870221, 0.021226706, 0.018695572, 0.0051325774, -0.0010362249, 0.0150662735, -0.008370153, -0.008845577, -0.007646971, 0.00176108, -0.002142759, -0.013117703, -0.018521473, 0.034766264, -0.0052999803, -0.001649757, -0.00096005644, -0.027561236, -0.019566068, 0.0021661955, -0.027119292, -0.04743533, -0.035435878, 0.032328878, -0.0009416421, 0.018936632, 0.028498694, -0.008129092, -0.019659813, 0.005879195, -0.014102032, -0.00030090698, 0.0034719391, 0.025632754, -0.008691566, 0.019150909, -0.013070829, -0.00196866, 0.044730093, -0.013010564, 0.016405499, -0.026114875, 0.04012316, 0.006501934, 0.019753559, -0.019900873, -0.046283595, -0.012736023, 0.023610525, -0.024762258, 0.016124262, 0.016874228, 0.0021829358, 0.07371091, 0.006542111, -0.042587336, 0.0032057683, 0.01391454, 0.018936632, -0.02184275, 0.0013425724, 0.018106313, -0.027963005, 0.0022448748, -0.006009769, 0.03144499, -0.019766951, -0.01861522, 0.018521473, -0.007841159, 0.02018211, -0.017342957, -0.02536491, 0.0050957487, -0.04250698, 0.008035347, 6.980576e-07, -0.047542464, 0.015200196, 0.047703173, -0.012059716, -0.0010688684, -0.057265233, 0.0033162544, -0.0016614752, -0.028498694, -0.0152537655, 0.009736161, 0.0026650564, -0.0068467846, -0.023864977, 0.0007918164, 0.029034384, -0.007707236, 0.041623093, -0.019887483, -0.0112494845, -0.014878782, -0.00737243, -0.027588021, -0.03353418, -0.030828943], 'node_index': 13814, 'node_id': 'Copper_(14012)'}"]
CPU times: user 5 ms, sys: 1.73 ms, total: 6.73 ms
Wall time: 12.5 ms

In [10]:

%%time

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_nodes')

# Load the collection into memory before query
collection.load()

# Query by expr on node_index
expr = "node_index in [13814, 13815]"
output_fields = ["node_index", "node_id", "node_name", "node_type", "desc", "desc_emb"]

results = collection.query(expr, output_fields=output_fields)

print(results)

%%time # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_nodes') # Load the collection into memory before query collection.load() # Query by expr on node_index expr = "node_index in [13814, 13815]" output_fields = ["node_index", "node_id", "node_name", "node_type", "desc", "desc_emb"] results = collection.query(expr, output_fields=output_fields) print(results)

data: ["{'node_type': 'drug', 'desc': 'Copper belongs to drug node. Copper is a transition metal and a trace element in the body. It is important to the function of many enzymes including cytochrome c oxidase, monoamine oxidase and superoxide dismutase. Copper is commonly used in contraceptive intrauterine devices (IUD). Copper is absorbed from the gut via high affinity copper uptake protein and likely through low affinity copper uptake protein and natural resistance-associated macrophage protein-2. It is believed that copper is reduced to the Cu1+ form prior to transport. Once inside the enterocyte, it is bound to copper transport protein ATOX1 which shuttles the ion to copper transporting ATPase-1 on the golgi membrane which take up copper into the golgi apparatus. Once copper has been secreted by enterocytes into the systemic circulation it remain largely bound by ceruloplasmin (65-90%), albumin (18%), and alpha 2-macroglobulin (12%).  Copper is nearly entirely bound by ceruloplasmin (65-90%), plasma albumin (18%), and alpha 2-macroglobulin (12%). Copper is incorporated into many enzymes throughout the body as an essential part of their function. Copper ions are known to reduce fertility when released from copper-containing IUDs. For use in the supplementation of total parenteral nutrition and in contraception with intrauterine devices.', 'desc_emb': [-0.008437114, -0.006676033, 0.017289387, -0.031070005, -0.02470869, 0.026931802, -0.025539009, -0.016030516, -0.022900736, -0.017262602, 0.01635193, 0.02225791, -0.013927933, 0.0072853803, -0.010954855, 0.01364, 0.033105627, 0.025713107, -0.0011316446, 0.019137517, -0.042480197, 0.02196328, 0.0012337605, -0.011530722, 0.001766102, 0.0112494845, 0.030427177, -0.03964104, -0.0038636622, -0.02178918, -0.00078553875, 0.0020724498, -0.02006158, 0.01814649, 0.010546392, 0.011095474, 0.019512499, 0.006207305, -0.002527786, -0.0065588513, 0.018280413, 0.006548807, -0.0063512716, 0.024239961, -0.0007826092, 0.009568758, 0.0057821013, 0.0011140673, -0.025391694, 0.0064517134, 0.009200471, 0.012454786, -0.0059997253, -0.015856417, -0.0007487101, -0.021869535, 0.0077942857, 0.018454513, 0.011999451, -0.02035621, -0.0113432305, -0.010820933, -0.011624468, 0.024373883, 0.00061897276, 0.03139142, 0.0070309276, 0.008109003, -0.0015032793, 0.017235817, 0.0064718015, 0.031150358, 0.012809681, -0.012829769, 0.037096515, -0.025820246, -0.065086305, 0.00038105118, 0.003024973, 0.01802596, 0.019057162, 0.0065119783, -0.0011835395, 0.01903038, 0.01008436, 0.015923379, 0.008470594, 0.025418477, -0.004456269, -0.0039708004, -0.0071246736, 0.01754384, 0.019887483, 0.03682867, -0.006200609, 0.03460556, -0.016418891, 0.050461974, -0.013064133, -0.011738301, -0.00844381, 0.01552161, -0.009287521, -0.01676709, -0.019485714, 0.005273196, -0.0142895235, -0.011678036, 0.016981365, -0.022606106, -0.011236092, 0.071407445, 0.010526303, -0.0386768, -0.008309888, 0.0015719145, 0.027427314, -0.007051016, -0.00069723366, -0.0017995827, -0.016539421, 0.014959136, 0.01676709, -0.023329288, 0.009401355, 0.011992754, -0.022351654, -0.019244654, 0.007546529, -0.007821071, 0.0013810751, -0.0016715194, 0.003853618, 0.04004281, -0.013372155, 0.0042955624, -0.020275857, -0.0016430608, -0.005882543, -0.024119431, 0.03361453, 0.021280276, -0.011952577, -0.0022632892, 0.0011249485, 0.040551715, 0.038891077, 0.008972803, -0.009321001, 0.004888169, 0.009890172, -0.0013693569, 0.030534314, -0.028739754, 0.021574905, 0.0130507415, -0.0073791263, 0.016218007, 0.014061855, -0.029355798, -0.0016413869, -0.035221603, -0.009990614, -0.0026014433, 0.017342957, 0.01982052, 0.012829769, -0.0066392045, 0.025726499, -0.005929416, 0.01220703, 0.02304805, -0.029543288, 0.031284284, 0.00592272, 0.009622327, 0.007881336, 0.014222562, -0.013767227, -0.029329013, -0.0033681493, -0.0037900049, 0.018722357, -0.009321001, -0.009856692, 0.0010152995, 0.018333983, 0.012896731, 0.011557506, -0.015856417, -0.027454099, 0.008450506, -0.029302228, -0.027748728, -0.63554233, 0.0041382033, -0.003960756, -0.021936495, -0.01101512, 0.019606246, 0.017704546, -0.004278822, 0.008852273, 5.335972e-05, 0.000274541, 0.023623917, 0.020757977, -0.023155188, -0.012046323, -0.02494975, -0.015267157, -0.0025411781, 0.006274266, -0.053033285, -0.052872576, 0.0046169762, -0.020570487, 0.016981365, 0.014329701, -0.002574659, 0.018575042, -0.019726776, -0.009347786, 0.0075063524, 0.002807349, 0.023356073, 0.019258047, 0.042399842, 0.051988687, 0.0048781247, -0.014878782, 0.020275857, -0.0076871477, 0.03503411, -0.010037486, -0.024963142, 0.015789455, -0.040016025, -0.0047944235, 0.011537418, 0.019539284, 0.018842887, 0.019673206, -0.006160432, -0.012963692, -0.015280549, -0.011959273, 0.0075800098, -0.00516271, 0.011617771, 0.029114736, -0.026034521, 0.008102308, -0.004181728, 0.001487376, -0.0014020005, 0.00071020745, -0.01766437, -0.019766951, -0.010995032, -0.036266197, -0.008731743, 0.0051392736, 0.028793324, 0.011075386, 0.012474875, 0.009749553, -0.016325146, 0.030507531, 0.015481433, -0.008169269, 0.007760805, -0.0039473637, 0.04620324, 0.013459205, 0.012059716, -0.008216142, -0.017744724, 0.030695021, 0.007519745, -0.006481846, -0.00019094409, 0.020369602, -0.018200058, 0.009454924, 0.0076201865, -0.0066392045, -0.0193384, -0.025391694, 0.018936632, -0.019231262, -0.022177555, 0.019807128, 0.0041482477, -0.0082094455, -0.027333569, -0.035194818, 0.015401079, 0.016151046, 0.014128816, -0.026851447, -0.0047475505, 0.038408954, -0.024333706, -0.0041348552, -0.0017008149, 0.0075264405, -0.016807266, -0.0034551988, -0.0041214633, 0.003189028, 0.012836466, -0.0002423159, -0.027855866, 0.030427177, -0.011510633, 0.0033396906, 0.007539833, -0.010057575, 0.0055443887, 0.0026717526, -0.02524438, -0.0032911438, -0.014008286, 0.002204698, -0.0062943543, 0.0224454, -0.010747275, 0.03179319, 0.013619912, 0.03969461, -0.0013752161, -0.0063613155, -0.03964104, -0.016700128, -0.011611075, 0.006870221, -4.9462367e-05, -0.012394521, -0.015079666, -0.023623917, 0.02453459, 0.003016603, -0.012481571, -0.017168857, -0.006826696, -0.0050220913, 0.03755185, 0.0193384, -0.012173549, -0.009200471, -0.0056950515, -0.01262219, -0.0032057683, -0.008470594, 0.038408954, -0.020329425, 0.005705096, -0.009374571, 0.0071849385, 0.0056749634, 0.022485577, -0.013593127, -0.03806076, 0.006421581, 0.0027654984, -0.013994894, -0.008504075, -0.025391694, 0.0005382008, -0.0013676828, -0.005279892, -0.02006158, 0.0006177172, -0.0077273245, -0.011323142, -0.014945744, -0.009622327, 0.022498969, 0.014557368, 0.01921787, 0.0050890525, 0.004061198, 0.027079117, 0.0032308786, 0.014463623, -0.021601688, -0.008852273, -0.010499519, -0.0053535495, -0.0070778006, 0.0051928423, -0.002079146, 0.02213738, 0.014222562, -0.012253903, -0.006615768, -0.034391284, 0.00048756137, -0.009957133, -0.0130172605, 0.005427207, 0.03002541, -0.0023955377, -0.005718488, -0.021333843, 0.0013534536, -0.013077525, -0.0031655915, 0.042239137, 0.0065387627, -0.018950025, -0.00015662647, -0.025083672, -0.0006281799, -0.010887894, 0.036266197, -0.0013852601, -0.0011827025, 0.0016397128, 0.02035621, -0.0025880511, -0.0035589887, -0.0057620127, 0.01101512, -0.009354482, -0.009287521, -0.010111144, -0.0024323664, 0.0072050267, 0.007345645, -0.011336534, 0.032409232, -0.015320727, -0.0017995827, 0.010532999, 0.028257633, -0.00056498527, 0.008745135, -0.013820795, 0.020811547, 0.018333983, 0.0041382033, 0.010298636, 0.0005545226, 0.011303053, -0.03790005, -0.016137654, 0.0016070692, -0.01020489, 0.012595405, 0.030105762, 0.035382308, 0.028043358, 0.030802159, -0.013251625, 0.0038435739, -0.013070829, 0.0035589887, 0.021052606, 0.017342957, -0.016378714, -0.019164301, 0.00043692195, -0.036667965, -0.00024566398, 0.009394659, -0.019740168, 0.013613216, 0.03144499, -0.01790543, 0.01724921, 0.013646696, -0.011202612, -0.028686184, -0.027266607, 0.0076268828, 0.019646421, -0.01032542, -0.026195228, -0.029114736, 0.015976947, 0.004928346, 0.01856165, -0.009508493, 0.019258047, -0.010466038, 0.025324732, -0.017624194, -0.0038804025, 0.021119568, -0.0087987045, 0.010921375, -0.012267295, -0.016619775, -0.021976672, -0.041676663, 0.0151600195, 0.053568974, 0.007713932, 0.011912401, -0.0077474127, 0.009762946, 0.01110217, -0.016392106, 0.0046872855, 0.0025227638, -0.04087313, 0.00070727785, 0.01821345, -0.020048188, 0.014075248, 0.030962866, -0.022003457, -0.018347373, -0.0094817085, -0.0050220913, 0.001211998, 0.012736023, 0.020034797, -0.01343242, 0.0070577124, 0.019258047, 0.0061738244, 0.0067563867, -0.01581624, 0.019311616, 0.0057352283, 0.010452646, -0.021467766, -0.012140069, -0.0019602897, -0.0013844232, 0.0045500146, -0.00853086, -0.023141798, 0.0020172067, -0.0075733135, 0.021507943, -0.021601688, 0.011229396, 0.023650702, 0.013901149, -0.0015576853, 0.006702818, 0.02871297, 0.03701616, 0.00019502455, 0.024963142, -0.0034250661, 0.01802596, 0.010198194, -0.0448908, -0.0026667304, 0.009776338, 0.016057301, -0.0027956308, 0.022271302, 0.017101895, 0.026971979, 0.009454924, -0.009756249, -0.01032542, -0.020865116, 0.012675758, -0.01163786, 0.0040310654, -0.010211586, 0.018066136, -0.000767543, 0.011651251, -0.034525204, 0.028016573, -0.028793324, 0.011216004, 0.010419166, -0.018427728, -0.0053736377, -0.0022381789, -0.011557506, 0.02938258, -0.007901424, 0.019807128, -0.015119842, -0.012494964, -0.029007599, -0.0042587337, 0.0060231616, 0.011216004, -0.030828943, 0.00090062834, -0.010332116, 0.017048327, 0.02686484, -0.0006809119, -0.020624055, 0.0066860775, 0.03026647, -0.0037866568, -0.02208381, 0.0014647766, -0.03800719, 0.011242788, -0.008276407, 0.037444714, -5.63939e-05, -0.013727049, -0.0001843526, 0.008510771, -0.022539146, -0.009756249, -0.00853086, -0.0043123025, 0.008571036, 0.011416888, 0.010405773, 0.008423721, -0.012501659, -0.00015045349, -0.011115562, -0.0003496631, -0.03313241, -0.011055297, -0.005514256, -0.0120864995, 0.006100167, -0.015776064, -0.02620862, -0.00028395743, -0.0097294655, -0.009990614, -0.016914405, 0.00078302773, 0.009079942, -0.011115562, 0.03356096, -0.01196597, -0.010767364, -0.0055912617, 0.007821071, 0.025940776, 0.023128405, -0.009769642, 0.020878509, -0.009716073, -0.013820795, -0.0048781247, 0.0052397153, 0.020985646, 0.032007463, -0.020691017, -0.032650292, -0.0014061856, -0.0060399016, -0.0091402065, -0.015468041, -0.0065722433, -0.0004624509, -0.007258596, 0.009033068, -0.011751694, -0.012320864, 0.0021193225, -0.025230987, 0.006813304, 0.01474486, -0.0061302995, 0.00942814, 0.012722631, -0.027775513, -0.031177143, 0.01957946, -0.015655532, -0.018066136, -0.007091193, -0.022565931, 0.04012316, 0.025177417, 0.025806854, 0.0005172754, 0.0060131173, -0.005109141, -0.00176108, 0.016365321, -0.010820933, -0.00090397644, 0.0082697105, 0.015843024, 0.015173412, -0.010352205, 0.01420917, -0.0014145557, -0.008021954, 0.015548394, -0.017450094, -0.0012814703, -0.022820383, -0.0072518997, -0.04269447, -0.004000933, -0.02130706, -0.0032660333, -0.031284284, -0.01844112, 0.014811821, 0.008792008, 0.019766951, -0.019900873, 0.028311202, -0.02375784, 0.035489447, 0.007218419, 0.0032040942, -0.005293284, -0.026971979, -0.00540377, 0.014276131, 0.020637447, -0.002810697, 0.015293942, -0.0067965635, -0.03098965, 0.0056548747, 0.012977084, -0.014235955, -0.007432695, 0.020034797, -0.022887344, -0.0052664997, -0.0039841924, -0.01569571, -0.046953205, -0.00029965147, 0.023128405, -0.0059093274, 0.018293805, -0.008919234, -0.0006009769, 0.02208381, -0.006073382, 0.042426627, -0.021588298, 0.038730368, 0.0048111635, -0.0042955624, -0.0052865883, -0.015762672, 0.0068467846, -0.002420648, 0.021333843, 0.04004281, -0.0011885617, -0.003135459, -0.011082081, -0.0072452035, 0.0050086994, -0.014557368, 0.007941601, 0.0017393176, 0.005105793, -0.02817728, -0.000826971, 0.0025177419, 0.00901298, 0.00045073268, -0.012997172, -0.021815965, 0.009274129, -0.0037129994, 0.018106313, 0.0016405499, 0.02627558, 0.0027839127, -0.00083534117, -0.00031513625, -0.027909435, -0.011631164, -0.0077206283, 0.017557232, 0.024293529, 0.014423446, 0.013405636, -8.83051e-05, -0.0039339717, 0.011216004, -0.020074973, 0.002909465, 0.036159057, 0.0059528523, 0.013927933, 0.025030103, 0.009274129, 0.0033296465, -0.029007599, -0.0013224841, -0.0047944235, -0.024132824, -0.015240373, -0.0040879827, 0.00015348766, -0.029462935, -0.016981365, -0.0016397128, 0.0062374375, -0.049819145, 0.0063043986, 0.049229886, -0.040149946, -0.014302916, -0.0050455276, 0.001155918, 0.029570073, -0.0095419735, -0.009334394, -0.029409366, 0.016070694, -0.010245066, 0.030828943, 0.026730917, 0.024333706, -0.004007629, 0.004720766, -0.015668925, -0.018722357, 0.013131095, -0.0072050267, -0.030507531, 0.0042754738, -0.013124399, -0.0040277173, 0.035060894, -0.0087049585, 0.015494825, 0.0010822607, 0.013994894, 0.0012019539, -0.009796427, 0.007546529, -0.029543288, -0.0004406885, 0.018963417, -0.011959273, -0.026222013, -0.012588709, 0.0012873295, 0.0110017285, -0.018588435, -0.0029914924, -0.023998901, 0.0077273245, -0.021561513, -0.020691017, 0.014316308, -0.00514597, -0.018133098, 0.015146627, -0.005879195, 0.003413348, -0.023798017, -0.015856417, 0.01850808, -0.00082320446, -0.0021025822, 0.00571514, 0.0059762886, -0.02166865, -0.02823085, 0.019043772, 0.0062608737, 0.01873575, 0.000115717354, -0.026530035, -0.0047676386, 0.018226843, -0.020436564, -0.027092509, -0.0053669414, 0.024266746, -0.0014061856, 0.023195365, 0.0076737553, -0.01562875, -0.008865666, 0.025324732, -0.039078567, 0.0077407164, 0.03425736, 0.00061939127, 0.033694886, 0.013345371, -0.038998213, -0.024936358, 0.020945469, 0.00045575478, 0.004734158, 0.014142209, -0.0091402065, 3.59655e-05, -0.012133373, -0.018066136, 2.0389169e-05, -0.011912401, 0.03198068, -0.007559921, -0.018494688, 0.0070711044, 0.002855896, -0.0065555032, -0.0012019539, -0.007586706, 0.014222562, -0.013994894, 0.0025512225, -0.0067362986, 0.03664118, -0.009602238, -0.0117784785, 0.01599034, -0.0017142071, -0.001816323, 0.023382857, -0.017222425, -0.023423035, -0.011269573, -0.026489858, -0.00024524546, -0.008604516, 0.005926068, -0.008490683, 0.0024691948, -0.001923461, -0.02782908, 0.010178105, -0.018575042, -0.01469129, -0.028766539, 0.012977084, -0.008450506, 0.0042085126, -0.0014672877, -0.0024457586, -0.0030048846, -0.015601964, -0.004760943, 0.0046605007, -0.023007875, -0.025913991, -0.009187079, 0.0075733135, 0.01707511, -0.019499106, -0.006809956, 0.00847729, 0.017235817, 0.19231261, -0.0063881003, 0.028391555, 0.027320176, -0.009441532, 0.0026449682, 0.015401079, 0.0048145116, 0.020503525, 0.015950162, 0.013218144, 0.007432695, -0.014543977, 0.0008219489, -0.03382881, -0.015642142, -0.042426627, 0.0019619637, -0.012709239, 0.017958999, -0.013211448, -0.020449957, 0.023409642, -0.014651114, 0.008062131, -0.0011977688, 0.010720491, -0.0028960726, 0.019016987, 0.00032894698, 0.0017778203, 0.005390378, 0.0108477175, -0.00043273685, 0.014168994, -0.0067329505, 0.011450368, -0.028552262, 0.01224051, -0.019445539, -0.0039774966, 0.010633442, -0.028257633, -0.010834325, -0.01023837, 0.013265017, 0.00040888193, -0.023141798, -0.014677899, -0.0045801476, -0.035998352, 0.010961551, 0.0021544772, 0.0027219735, -0.0039339717, 0.012695847, -0.003026647, 0.0005666593, 0.008671478, 0.0038770544, -0.0035824252, -0.0019586156, -0.012508355, 0.040685635, -0.0028207414, 0.030373607, -0.023208758, -0.009488405, 0.02679788, -0.024976535, -0.0007658689, 0.008229534, -0.0071380655, 0.00011770526, -0.0041984683, 0.007425999, -0.01044595, 0.017932214, 0.0131110065, -0.00018236469, -0.010767364, -0.01110217, -0.029864702, 0.013680177, -0.006200609, -0.0011400147, 0.014651114, -0.0065320665, -0.025873814, -0.009441532, 0.012106588, -0.019847305, -0.00853086, -0.012715935, 0.013305194, 0.0055778692, 0.011416888, 0.0060131173, -0.002914487, 0.0052899364, -0.023838194, 0.072318114, 0.009086638, 0.014530584, 0.01023837, 0.01706172, -0.021119568, 0.01880271, 0.0057553165, -0.021923102, 0.017209033, -0.02662378, -0.0134525085, -0.011416888, 0.0065655475, 0.025646146, 0.018775925, -0.011905705, 0.010419166, -0.009146903, 0.0070108394, -0.024400668, 0.00925404, 0.004181728, -0.0052999803, 1.2934501e-05, -0.027373746, 0.015414472, -0.020543702, -0.035757292, -0.011631164, -0.008135788, 0.0059327637, 0.009930349, -0.012454786, 0.020798154, 0.01730278, -0.004847992, 0.004690633, 0.024065861, -0.01826702, -0.003512116, 0.028980814, 0.0059695924, 0.005343505, -0.020744586, -0.02154812, -0.00063445757, -0.02220434, -0.019793736, 0.001487376, -0.017731331, -0.0049417377, -0.0005821441, 0.013847579, -0.022632891, -0.009508493, -0.036185842, 0.01396811, -0.017570624, -0.016606383, 0.001157592, 0.028820107, 0.0021193225, -0.013847579, 0.009180383, -0.16874227, 0.015588571, 0.011082081, -0.027400529, 0.013425724, 0.0067496905, -0.02381141, 0.01844112, 0.0076871477, -0.011162435, 0.033587746, 0.008571036, -0.022177555, -0.022659676, -0.003512116, -0.008149181, -0.006542111, 0.005323417, 0.021333843, 0.01635193, 0.023061443, -0.027668376, 0.0059662443, -0.009829907, 0.03431093, 0.027146077, 0.017383132, -0.0064483653, 0.009006284, -0.009448228, -0.00042018163, 0.019191084, 0.0417838, 0.00017807499, 0.025043495, 0.004837948, 0.017275995, -0.026637172, 0.00018121379, 0.022780206, 0.0069103977, 0.0014488733, -0.014543977, 0.02698537, -0.020945469, 0.03187354, -0.012896731, 0.002855896, 0.029329013, 0.002906117, 0.026945194, -0.003800049, -0.0041448995, 0.024065861, 0.01999462, 0.0019000245, -0.0012990476, 0.036453687, -0.0031388071, -0.011122258, 0.0028358076, -0.017450094, 0.017195642, 0.0009968851, 0.023530172, -0.015079666, -0.0011492219, 0.019070555, -0.025391694, -0.002743736, -0.012756112, 0.008169269, 0.018789317, -0.010365597, 0.019793736, 0.019043772, -0.019766951, 0.0050823563, -0.003356431, 0.015950162, -0.0072719883, 0.009937045, -0.038757153, -0.012006146, -0.016365321, -0.01574928, 0.01683405, -0.0051359255, -0.004653805, 0.0034418066, 0.017329564, -0.02137402, -0.031311065, -0.03157891, 0.026596995, 0.01310431, 0.014343092, -0.009655808, -0.01844112, -0.022070417, 0.025592577, -0.021226706, -0.029811133, -0.0023938636, 0.044676524, 0.012494964, -0.01184544, -0.017048327, 0.02782908, -0.0025076976, 0.009849995, -0.0013316913, 0.0193384, 0.040149946, -0.020195503, 0.034926973, 0.020141935, -0.020824939, -0.010004006, -0.020262465, 0.045640767, 0.007499656, -0.00045240673, 0.010720491, -0.026288973, -0.0073992144, -0.11935167, -0.0065588513, 0.02513724, 0.013927933, -0.009361179, 0.0033095581, -0.01903038, 0.029275443, 0.013666784, 0.022485577, -0.024427451, -0.016726913, -0.009896868, -0.005882543, 0.008872362, 0.009207168, -0.009501797, -0.0030015365, -0.027775513, 0.032141387, -0.001972008, -0.023074836, 0.006398144, 0.01861522, -0.008109003, -0.010499519, -0.005038832, 0.0130172605, 0.0058658025, 0.0071447617, 0.02441406, 0.00042562225, 0.035194818, -0.0025076976, -0.016874228, 0.007004143, 0.018695572, 0.00035803326, 0.044542603, -0.02016872, -0.004563407, 0.010620049, 0.0044730096, -0.011510633, 0.026061306, -0.010432558, -0.037096515, -0.005440599, -0.014450231, -0.011765086, -0.025954168, -0.0095419735, -0.03736436, -0.0046337163, 0.011818655, 0.005946156, 0.021226706, 0.019191084, -0.029757565, -0.013171271, -0.0018447816, -0.0039071874, -0.012227118, 0.007452783, 0.031016435, -0.014356485, -0.008417025, -0.002807349, 0.00208919, -0.0405785, -0.025030103, -0.0056816596, -0.016606383, -0.0039205793, -0.019164301, 0.04028387, 0.019351792, 0.012890034, 0.01969999, -0.041703448, -0.011865527, -0.024373883, -0.0050723124, -0.033802025, -0.0056247422, 0.038542878, 0.0364269, 0.0030567795, 0.009742857, -0.019432146, -0.019351792, 0.036989376, 0.0048982133, -0.015334118, 0.0010881198, 0.043015886, 0.0024256702, -0.00086128863, 0.018226843, 0.007981777, -0.016485853, -0.024132824, -0.049712006, 0.0054707313, 0.025913991, -0.0067999116, 0.00084998895, 0.00028479443, 0.0081156995, 0.004071242, -0.0093879625, 0.012970388, -0.010365597, -0.004821208, 0.005705096, -0.011979362, -0.02494975, 0.0039272755, 0.033052057, -0.0040879827, 0.0024859353, 0.008182661, -0.014932351, 0.0040411097, -0.0035991655, 0.013445812, -0.042828396, -0.0010069293, 0.0075532254, -0.013767227, -0.0077808932, -0.006106863, 0.0077675013, -0.035596583, -0.0016296686, -0.0072719883, -0.005122533, -0.028445125, -0.01486539, -0.003245945, 0.014396662, 0.02161508, -0.00023792157, -0.023315895, -0.0058591063, -0.011651251, 0.004235297, -0.007024232, -0.026637172, 0.010064271, -0.006763083, 0.013044045, 0.026596995, 0.012126677, 0.014905566, -0.032462798, -0.0009801448, -0.0042520375, -0.003525508, -0.0138877565, -0.0343645, -0.02058388, 0.011564203, 0.011021816, -0.0034016299, 0.00616378, 0.029409366, -0.001814649, -0.027614806, -0.029275443, 0.030400392, -0.007218419, -0.011724909, -0.016445676, -0.004653805, 0.0107539715, 0.013660088, 0.0042520375, -0.021467766, 0.002519416, -0.004667197, 0.013144487, -0.017958999, -0.014597545, -0.013633303, 0.022726638, 0.004114767, 0.0128632495, 0.006870221, 0.021226706, 0.018695572, 0.0051325774, -0.0010362249, 0.0150662735, -0.008370153, -0.008845577, -0.007646971, 0.00176108, -0.002142759, -0.013117703, -0.018521473, 0.034766264, -0.0052999803, -0.001649757, -0.00096005644, -0.027561236, -0.019566068, 0.0021661955, -0.027119292, -0.04743533, -0.035435878, 0.032328878, -0.0009416421, 0.018936632, 0.028498694, -0.008129092, -0.019659813, 0.005879195, -0.014102032, -0.00030090698, 0.0034719391, 0.025632754, -0.008691566, 0.019150909, -0.013070829, -0.00196866, 0.044730093, -0.013010564, 0.016405499, -0.026114875, 0.04012316, 0.006501934, 0.019753559, -0.019900873, -0.046283595, -0.012736023, 0.023610525, -0.024762258, 0.016124262, 0.016874228, 0.0021829358, 0.07371091, 0.006542111, -0.042587336, 0.0032057683, 0.01391454, 0.018936632, -0.02184275, 0.0013425724, 0.018106313, -0.027963005, 0.0022448748, -0.006009769, 0.03144499, -0.019766951, -0.01861522, 0.018521473, -0.007841159, 0.02018211, -0.017342957, -0.02536491, 0.0050957487, -0.04250698, 0.008035347, 6.980576e-07, -0.047542464, 0.015200196, 0.047703173, -0.012059716, -0.0010688684, -0.057265233, 0.0033162544, -0.0016614752, -0.028498694, -0.0152537655, 0.009736161, 0.0026650564, -0.0068467846, -0.023864977, 0.0007918164, 0.029034384, -0.007707236, 0.041623093, -0.019887483, -0.0112494845, -0.014878782, -0.00737243, -0.027588021, -0.03353418, -0.030828943], 'node_index': 13814, 'node_id': 'Copper_(14012)', 'node_name': 'Copper'}", '{\'node_type\': \'drug\', \'desc\': "Oxygen belongs to drug node. Oxygen is an element displayed by the symbol O, and atomic number 8. It is an essential element for human survival. Decreased oxygen levels may be treated with medical oxygen therapy. Treatment with oxygen serves to increase blood oxygen levels and also exerts a secondary effect of decreasing blood flow resistance in the diseased lung, leading to decreased cardiovascular workload in an attempt to oxygenate the lungs. Oxygen therapy is used to treat emphysema, pneumonia, some heart disorders (congestive heart failure), some disorders that cause increased pulmonary artery pressure, and any disease that impairs the body\'s ability to take up and use gaseous oxygen. Higher level of oxygen than ambient air (hyperoxia) can be introduced under normobaric or hyperbaric conditions. Oxygen therapy increases the arterial pressure of oxygen and is effective in improving gas exchange and oxygen delivery to tissues, provided that there are functional alveolar units. Oxygen plays a critical role as an electron acceptor during oxidative phosphorylation in the electron transport chain through activation of cytochrome c oxidase (terminal enzyme of the electron transport chain). This process achieves successful aerobic respiration in organisms to generate ATP molecules as an energy source in many tissues. Oxygen supplementation acts to restore normal cellular activity at the mitochondrial level and reduce metabolic acidosis. There is also evidence that oxygen may interact with O2-sensitive voltage-gated potassium channels in glomus cells and cause hyperpolarization of mitochondrial membrane.  Oxygen binds to oxygen-carrying protein in red blood cells called hemoglobin with high affinity. The amount of oxygen molecules bound to the fixed amount of circulating hemoglobin in the blood determines the overall oxygen saturation level and this oxygen-delivering capacity is regulated by Bohr effect. Oxygen therapy improves effective cellular oxygenation, even at a low rate of tissue perfusion. Oxygen molecules adjust hypoxic ventilatory drive by acting on chemoreceptors on carotid bodies that sequentially relay sensory information to the higher processing centers in brainstem. It also attenuates hypoxia-induced mitochondrial depolarization that generates reactive oxygen species and/or apoptosis. Oxygen therapy in clinical settings is used across diverse specialties, including various types of anoxia, hypoxia or dyspnea and any other disease states and conditions that reduce the efficiency of gas exchange and oxygen consumption such as respiratory illnesses, trauma, poisonings and drug overdoses. Oxygen therapy tries to achieve hyperoxia to reduce the extent of hypoxia-induced tissue damage and malfunction. ", \'desc_emb\': [0.0011131193, 0.00035166115, -0.011181605, -0.04077155, -0.011246652, 0.02419754, -0.016274797, -0.01151985, -0.027632028, 0.0079422565, -0.019462107, 0.01414125, -0.00018365656, 0.023781238, -0.0011261287, -0.010680742, 0.035047404, 0.0017481421, 0.010882388, -0.014713665, -0.018174173, 0.017953012, -0.016600031, 0.009880662, -0.017913984, 0.01151985, 0.033954613, -0.0328358, 0.019579193, -0.024535784, 0.021543616, 0.001392822, -0.020190636, -0.010231917, -0.0010399413, 0.009100096, -0.0030637207, -0.003120637, 0.013894071, -0.004823246, 0.027762122, 0.017744862, -0.003639388, 0.0068234457, -0.0026230263, -0.01823922, -0.0025270816, -0.037831422, 0.0065014623, -0.004302869, 0.0029531405, 0.009568435, 0.004058942, -0.0051192106, 0.0036361355, -0.032263387, -0.016105674, 0.02557654, 0.010485601, -0.008527681, -0.003802006, -0.013568835, -0.030988462, 0.012612642, -0.0056265783, 0.022467285, 0.0030474588, 0.04420604, -0.01037502, 0.006468939, 0.028100368, 0.03715493, 0.012924869, 0.024210548, 0.008384577, -0.005893272, -0.024392681, 0.005278576, 0.017419627, 0.0045435433, 0.010335992, -0.021556625, -0.00673238, 0.026435161, 0.029895669, -0.0028344295, -0.0046281046, 0.012261388, 0.004631357, 0.010023765, 0.010745789, 0.004556553, 0.009236695, 0.0147917215, 0.006397387, 0.03699882, 0.002775887, 0.032185327, 0.0026799424, -0.018187182, 0.00010651471, -0.0014310373, -0.034813233, -0.010179879, -0.031612914, 0.014921816, -0.005350128, 0.0019855641, 0.003447499, -0.007090139, -0.011428784, 0.056721114, -0.008638261, -0.013321656, 0.0046281046, 0.02326086, 0.026851462, 0.007623526, -0.017575739, 0.0045012627, 0.0059843375, 0.017133418, 0.010257935, -0.005460708, -0.00026181477, -0.01346476, -0.032341443, -0.018889692, -0.005050911, -0.0074088704, 0.015090939, 0.015806457, 0.016769154, -0.011448299, -0.021881862, 0.042046476, -0.011832076, -0.018174173, -0.029505387, -0.024626851, 0.014388429, 0.0074153747, -0.0053826515, -0.0074218796, -0.011227137, 0.020476842, 0.03533361, 0.01528608, 0.013646891, -0.003623126, 0.0057176445, -0.009535912, 0.0016074777, -0.028230462, 0.00014625445, 0.009340771, -0.009568435, 0.017172446, -0.014037175, -0.020398786, 0.01639188, -0.016508967, -0.016105674, -0.0051192106, 0.020177625, 0.007838181, -0.0015351127, 0.0139461085, -0.008124389, 0.0012342697, -0.00445573, 0.03759725, -0.003967876, 0.008566709, -0.002275024, 0.035229534, -0.0039386046, 0.03174301, 0.002837682, -0.017432636, -0.007486927, 0.017979031, 0.0079227425, 0.009666006, -0.015871504, 0.005633083, 0.027085632, -0.009692025, 0.027189707, -0.04425808, -0.00073828513, 0.025615567, -0.013829024, -0.009581445, -0.64610034, -0.015689373, -0.038924213, -0.011233643, 0.014583571, 0.018161165, 0.014609589, -0.0010496983, -8.110566e-05, 0.022779511, 0.0058022058, 0.019696277, -0.0050931918, -0.0045142723, 0.016118683, -0.03585399, -0.018382324, -0.013568835, 0.032133292, -0.038429856, -0.023364935, -0.016261786, -0.042983156, 0.00016800458, 0.012801278, 0.026396133, 0.021582644, -0.005714392, 0.003805258, 0.028178426, 0.007207224, 0.010069299, 0.032185327, 0.018811636, 0.053286623, -0.0018424605, -0.020021513, 0.017237494, 0.008579719, 0.018538438, 0.021998946, -0.0034052182, 0.0255245, -0.025732651, -0.013438741, 0.019449098, 0.027996292, -0.015585297, 0.0085927285, 0.0087293275, 0.02054189, 0.005174501, -0.022402238, 0.012957392, 0.00079682755, -0.00049191905, 0.021517597, -0.020932171, 0.0094188275, -0.010550647, -0.00042361955, 0.010479095, -0.009945709, -0.02624002, -0.0026978306, 0.024991114, -0.028022312, 0.0030068045, 0.010270945, 0.015624325, 0.011350728, 0.03556778, -0.0048557697, 0.00478747, 0.008996021, 0.0072657666, -0.00083585584, 0.0035938548, -0.012989916, 0.03806559, 0.014180278, -0.0053989133, -0.005805458, -0.003057216, 0.04496059, 0.015208024, -0.02327387, 0.016053636, 0.02374221, -0.0009505015, 0.015572287, 0.012807784, -0.0036328833, -0.028620746, -0.012664679, -0.0011692225, 0.009080582, 0.0037076874, 0.027033595, 0.015910532, -0.010615694, -0.005167996, -0.010498609, 0.018811636, -0.012092265, 0.0040361756, -0.037363082, -0.023442993, 0.035932045, -0.02236321, -0.006267293, -0.019110853, 0.023143776, -0.0022864072, 0.041265912, -0.006758399, -0.0039548664, 0.0010732779, 0.004361411, -0.05151734, 0.015936552, -0.016600031, -0.0014838881, -0.012846812, 0.021075277, -0.008722822, -0.0075584785, -0.013529807, -0.015116957, -0.014778712, -0.0017595254, -0.011240147, 0.007434889, 0.0124890525, 0.006059142, 0.00071755133, 0.030650217, 0.009679016, 0.01800505, -0.024080455, 0.0047842176, -0.00135542, -0.011031996, -0.0028815886, -0.010075803, -0.0034442465, -0.015299089, 0.008781365, -0.018863672, 0.0055029886, -0.017874956, -0.0064299107, -0.0117800385, 0.04680793, 0.0033824518, -0.023013681, 0.005668859, -0.015728401, -0.024587821, -0.007499936, -0.0029905427, 0.024314623, 0.0055094934, 0.0026457927, -0.021959918, -0.006133946, -0.018187182, 0.02167371, -0.015884513, -0.042931117, -0.012807784, -0.009412322, -0.006986064, -0.013972128, -0.004468739, 0.016001599, -0.011695477, -0.0068039317, 0.0011090538, 0.0027872704, 0.0087098135, -0.017003324, 0.0012049985, -0.010401039, 0.0013456629, 0.009191162, 0.019488126, 0.0139851365, 0.013737958, 0.038455874, -0.0125475945, 0.031404763, -0.015663354, 0.014180278, -0.002946636, -0.027241746, -0.0027466158, -0.029349273, 0.004123989, 0.025459453, 0.017224485, -0.026539236, 0.031352725, -0.018343296, 0.030702254, 0.0066868467, -0.009048059, -0.0031889365, 0.010179879, 0.01505191, -0.007825172, -0.01128568, 0.00038154217, 0.0049078073, -0.018915711, 0.02283155, -0.011558878, -0.006381125, 0.0048460127, -0.016899249, 0.003029571, -0.0058379816, 0.0066868467, -0.0036003597, -0.015364137, 0.0051289676, 0.028985009, 0.0027970274, -0.0032328432, -0.007454403, -0.007727601, -0.017953012, 0.0053728945, -0.009490379, 0.003603612, 0.01937104, -0.00041508212, -0.009490379, 0.03078031, -0.007278776, -0.009054563, 0.0053566326, 0.032809783, -0.022675436, -0.006751894, 0.011968675, 0.03124865, 0.0054574558, -0.015858496, -0.016756145, -0.017679814, -0.0075454693, -0.0065632574, -0.0012749241, 0.0020684993, -0.005441194, 0.02215506, -0.0069015026, 0.04514272, 0.030390028, 0.04077155, -0.011669459, 0.034110725, -0.012287406, -0.0009822119, 0.0066380617, -0.0029970473, -0.009048059, -0.005652597, -0.0016066646, -0.02258437, 0.0064624343, 0.008404092, -0.016040627, 0.03879412, 0.02143954, -0.027996292, 0.008644766, 0.033460252, 0.026604284, -0.016977306, -0.0018847411, 0.01826524, 0.031222632, -0.012671185, -0.009054563, -0.02072402, 0.011214129, -0.004494758, 0.0094773695, -0.004605338, -0.006875484, 0.005675364, 0.008878936, -0.0008216268, -0.046859965, 0.000192194, 0.018902702, 0.033746462, -0.003694678, -0.018681541, 0.008202446, -0.016404891, 0.004485001, 0.040511362, 0.008267493, 0.009275723, 0.019930447, -0.0049273213, -0.010030271, 0.011448299, -0.011162091, 0.002756373, -0.04334742, 0.018317277, 0.023182804, -0.0037434634, 0.011402765, 0.009139124, -0.032081254, 0.0030702255, -0.011669459, 0.0058802622, -0.014895797, 0.029739557, 0.008859422, -0.002883215, 0.02371619, -0.0073373183, 0.015025891, -0.0016700856, -0.017497683, 0.015494231, 0.0071877097, 0.0004768769, 0.001756273, 0.020333739, -0.0016326834, 0.004829751, 0.033642385, 0.0029254956, -0.03171699, -0.018512418, -0.006270545, 0.0052297907, -0.029817613, 0.015455202, 0.034839254, 0.02054189, 0.014310373, 0.0053631375, 0.024756944, 0.021608662, -0.0020424805, -0.0024051184, 0.025511492, 0.009236695, 0.012573614, -0.019462107, -0.014323382, -0.0037694823, -0.0021904628, 0.03101448, 0.012794774, 0.007929248, 0.023768228, -0.0039711283, 0.0038638005, -0.0040003997, -0.02327387, -9.086273e-05, 0.0034540035, -0.0019481621, -0.004859022, 0.013191562, 0.008150408, -0.009900176, -0.035047404, 0.009718044, -0.0041337465, 0.0004931387, 0.0145965805, -0.027658047, 0.0145965805, -0.013503788, -0.03392859, 0.03309599, -0.019097842, -0.0006447799, 0.0017416374, 0.014232316, 0.011890619, -0.005171248, -0.017718844, 0.006413649, -0.0058022058, -0.0060461327, -0.0052200337, 0.038611986, 0.01345175, 0.020711012, -0.00888544, 0.0035288078, 0.0005793262, -0.0035157984, 0.0077731344, -0.0012033723, -0.03582797, 0.010030271, -0.004715918, 0.004592329, 0.002275024, -0.01665207, 0.004241074, 0.022558352, -0.023117756, -0.0102058975, -0.015377146, 0.0070836344, 0.04129193, 0.0132175805, 0.012976906, -0.0145965805, -0.017315552, -0.013399713, -0.004712666, -0.00798779, -0.016990315, 0.0030507112, -0.025641587, -0.012892345, 0.014336391, -0.0023124262, -0.013334665, 0.01938405, 0.0027905228, 0.0022050983, 0.006986064, -0.009542417, -0.004550048, -0.012723222, 0.035697874, -0.017536711, -0.0005329801, -0.025654595, -0.013711939, 0.025654595, 0.021790795, -0.0068364553, -0.013126515, -0.005184258, -0.011305194, -0.0023872303, 0.00627705, 0.013568835, 0.024782963, 0.0037922487, -0.0230397, -0.034318876, 0.006771408, -0.0102254115, -0.011923143, 0.0012196341, -0.013568835, 0.0018538437, 0.013080982, 0.007350328, -0.0102449255, 0.026591275, -0.029037047, -0.030364009, 0.007948762, -0.002920617, -0.0077406107, -0.015455202, -0.023638133, -0.043945853, 0.009184658, -0.020268692, -0.014011156, -0.011071024, -0.010888892, 0.017757872, 0.030025763, 0.012437015, -0.023078728, 0.027944256, 0.012983411, -0.010667732, 0.01139626, -0.011018987, 0.011123062, -0.009854643, 0.012599633, 0.0013188309, -0.015377146, 0.014245326, -0.0044101966, -0.0006561631, 0.0035613314, -0.008052837, -0.0032409742, -0.0069210166, -0.015129967, -0.021270417, 0.016756145, -0.03215931, 0.009022039, -0.040043026, -0.010823846, 0.038846157, 0.00911961, 0.0030165615, -0.015325109, -0.004706161, -0.023560077, -0.0045337863, 0.014401439, -0.00069356523, -0.005233043, -0.02259738, -0.018174173, -0.03124865, -0.0075129457, 0.0010521376, -0.016574014, -0.012690699, -0.021179352, -0.014349401, 0.007890219, -0.009275723, -0.010784817, -0.00211078, -0.03650446, -0.00019341362, -0.009770081, -0.012280902, -0.025420425, -0.014037175, 0.034995366, 0.0032458527, 0.03304395, -0.011155586, -0.0042833546, 0.02463986, -0.02211603, 0.0069535403, -0.012853317, 0.04150008, -0.006179479, -0.014050184, -0.005906281, -0.02350804, 0.009191162, -0.01197518, 0.010869378, -0.017926995, -0.0025531005, -0.014999872, 0.012593128, 0.006189236, 0.01392009, -0.008839907, 0.02672137, -0.014115231, -0.0034832747, -0.018122135, -0.01662605, -0.024080455, 0.011539364, -0.00045126458, -0.02804833, -0.00039353524, -0.0063323397, -0.020619946, 0.005506241, -0.0020310972, -0.0007882901, 0.00684296, 0.0015188509, -0.021764776, -0.005067173, -0.026500208, -0.00991969, 0.0021888365, 0.025238294, -0.0155592775, 0.025459453, 0.01691226, -0.008137398, -0.0056135687, -0.021556625, -0.007090139, 0.032055236, -0.008345549, 0.005080182, 0.02577168, -0.0032263387, 0.0056460924, -0.0009456229, -0.0070771296, -0.007851191, -0.014973854, 0.009861148, 0.0028539437, 0.034318876, -0.019188909, -0.007857695, -0.01278827, -0.026083907, -0.025459453, -0.0027384849, 0.008013809, -0.03424082, -0.0056265783, -0.000110783425, -0.0029515144, 0.025303341, -0.017601758, -0.01939706, -0.0079422565, 0.030364009, -0.00809837, 0.025641587, 0.0042963643, 0.01662605, -0.006169722, -0.008117884, -0.014310373, -0.027632028, 0.008300016, 0.013373693, -0.041734252, -0.028126387, -0.02395036, -0.020216653, 0.0025238295, -0.0027954013, 0.020789068, 0.0022538838, 0.002884841, -0.014544542, -0.029271217, 0.025615567, -0.017068371, 0.0077796387, 0.023156784, -0.04402391, -0.021205371, -0.005460708, 0.0028246725, 0.0049891165, -0.014726674, -0.013659901, -0.040927667, 0.017458655, -0.012157312, -0.0020717517, -0.017835928, 0.017432636, -0.030650217, 0.021946907, 0.0026246523, 0.017406616, -0.013133019, -0.008260988, -0.0006988503, -0.00083748205, -0.030832348, -0.0019676762, 0.005704635, -0.007623526, -0.023573088, -0.008599233, 0.008423606, 0.012931373, 0.015702382, -0.023898322, -0.013672911, 0.0030897395, -0.047042098, -0.00087407103, -0.0038605484, 0.024444718, -0.0039808853, 0.031769026, 0.0011602786, 0.0037597253, 0.013269618, 0.009262714, -0.011767029, -0.017263513, 0.013776986, 0.0020993967, 0.00069315865, -0.014167269, -0.041604158, -0.0058542434, 0.0147917215, -0.013002925, 0.0132566085, 0.017172446, 0.010589676, 0.012300416, 0.0017156185, 0.020060541, -0.0035775932, 0.00014391681, -0.010134346, -0.023143776, -0.005294838, 0.01894173, 0.036556497, 0.0052720713, 0.010927921, -0.0006175414, 0.004839508, 0.0063225827, -0.00061103667, 9.599536e-05, 0.0029222432, -0.0057924488, -0.017237494, 0.007838181, -0.025199266, 0.0092432, 0.00644292, 0.0155202495, 0.0018343296, 0.0031190107, 0.0070771296, -0.0014481122, 0.025797699, 0.0052265385, -0.022376219, 0.014232316, 0.013829024, -0.02557654, 0.02600585, 0.0023156784, -0.007389356, -0.014440467, 0.027085632, 0.0016944782, -0.0018668532, -0.0074088704, -0.019475115, -0.016105674, -0.020919163, -0.017055362, 0.0065860236, -0.008260988, -0.0062380214, -0.0062347692, 0.0009456229, -0.013087486, -0.0016448798, 0.0029450096, 0.0013107001, 0.013763976, 0.19868001, -0.0058444864, -0.008254483, 0.017068371, -0.003185684, 0.012619147, -0.0106222, -0.02807435, -0.01128568, 0.013933099, 0.02371619, -0.008547195, -0.018850664, -0.0053111, -0.011058015, 0.0009813989, -0.0230397, -0.007324309, -0.01505191, 0.0030864873, -0.030416047, 0.00097408105, 0.011552374, 0.00205549, 0.012619147, -0.0044817487, 0.0117605245, -0.00014096937, 0.045715135, 0.008182932, -0.0055192504, 0.010017261, -0.0037011828, -0.00025002498, 0.012645165, -0.0032230862, 0.0118190665, -0.030598179, -0.0021498082, -0.009262714, 0.00216607, 0.022987662, -0.030598179, -0.0044004396, 0.029791594, 0.047796644, -0.014128241, -0.019683268, -0.02009957, -0.0115848975, -0.034006648, 0.0076950775, 0.0016879735, 0.028724821, -0.0016188609, 0.019266965, 0.009835129, 0.013107001, -0.0014798227, 3.1411928e-06, -0.0071812053, -0.01060919, -0.0070641204, 0.03231542, -0.0087683555, 0.0264872, -0.040173117, -0.0007886967, -0.016495956, 0.010693751, 0.0022977905, 0.00035897896, -0.0024262588, -0.009555426, -0.016482947, -0.010212402, 0.01619674, 0.0108563695, -0.009334266, 0.022480294, -0.008378073, -0.022063993, -0.025667604, 0.017380599, -0.006771408, -0.023364935, -0.003558079, 0.02259738, -0.009262714, 0.0063128257, -0.0032247123, -0.020112578, -0.017731853, 0.020463834, 0.021023238, -0.020789068, 0.008872431, 0.014583571, -0.020021513, 0.01937104, -0.01084336, 0.07873307, -0.014648618, -0.002457156, 0.020476842, 0.010745789, -0.029505387, 0.02737184, -0.0023807257, 0.010108327, 0.015767429, -0.043451495, -0.0044069444, -0.008085361, 0.010368516, 0.035723895, 0.008117884, -0.020177625, 0.016066646, -0.015025891, -0.0016302442, -0.018135145, -0.008781365, 0.022727475, -0.013295637, 0.0010846612, -0.02942733, 0.035099443, -0.014505514, -0.02713767, 0.021855842, -0.009054563, -0.0010066046, 0.0039093336, -0.012365463, 0.01849941, -0.0040556896, -0.022480294, 0.0073112994, 0.009913186, -0.013933099, -0.009535912, 0.010101822, 0.017926995, 0.016704109, -0.051673453, 0.0046508713, -0.02121838, 0.002483175, -0.012872831, -0.014882787, -0.02713767, 0.008917964, -0.019358031, -0.00058827014, -0.020203644, 0.016222758, -0.02305271, 0.016313825, -0.005916038, -0.007981285, 0.00398739, 0.02349503, -0.032445516, -0.0070185876, -0.0029889166, -0.16214953, 0.004452477, 0.020867126, -0.01782292, 0.0198654, -0.017003324, 0.016274797, -0.0031011228, -0.0027888964, 0.017848937, 0.018330287, 0.009802605, -0.0034767701, -0.0013066346, 0.011097044, -0.01983938, 0.018889692, 0.0282825, 0.027449897, -0.002475044, 0.016118683, 0.0045110197, 0.028932972, 0.0028132892, 0.012684194, -0.012137798, -0.005434689, 0.00991969, 0.014323382, 0.010088813, -0.023768228, 0.012749241, 0.031795047, 0.0021514345, -0.0076755635, 0.008527681, 0.032341443, -0.011116558, -0.025615567, 0.015129967, -0.0048362557, 0.014895797, -0.0040296707, 0.0024278848, -0.027970273, 0.016977306, 0.02443171, -0.008976506, 0.011968675, 0.02669535, 0.022675436, 0.011142576, -0.017900975, 0.024782963, 0.0044264584, -0.0020587423, 0.00043337664, 0.010628704, -0.021764776, -0.010739285, 0.012833802, -0.04496059, -0.007025092, 0.0017074876, 0.019943455, -7.404195e-05, -0.020060541, 0.0069405306, -0.008833403, 0.0062185074, -0.034266837, -0.025017133, 0.029271217, 0.013659901, 0.008761851, 0.011350728, -0.015012882, 0.0052558095, 0.0025075676, 0.033954613, 0.010368516, 0.004302869, -0.020658974, -0.008332539, -0.0034117228, 0.012859821, 0.01197518, -0.007525955, -0.015416174, -0.004413449, 0.01963123, -0.02487403, -0.015637334, -0.014570561, 0.016574014, 0.018967748, 0.019071825, -0.010862874, 0.0025108198, -0.0058900192, 0.019722296, -0.0006394948, -0.0023530805, -0.009692025, 0.050918907, 0.016313825, -0.0033889564, -0.009178152, 0.04220259, -0.0004467926, -0.010524629, -0.001056203, 0.017172446, 0.044180024, -0.007343823, 0.012716717, 0.023482021, -0.021907879, -0.011506841, -0.0058835144, 0.06551549, 0.013568835, -0.018681541, 0.03192514, -0.015103948, -0.04287908, -0.1313432, -0.030494103, 0.029947707, 0.012144303, 0.0064396677, 0.024496756, -0.010218907, 0.0046443664, -0.002757999, 0.025615567, -0.01153286, -0.029011028, -0.012391482, 0.006569762, 0.025953813, -0.007831677, -0.004946836, -0.0040166616, -0.00072486914, 0.028880933, -0.013503788, -0.00089602446, -0.0040394277, 0.0037239492, -0.004810237, -0.028568707, -0.0077861436, 0.017640786, 0.0046541234, -0.009548921, 0.01937104, -0.0030994967, 0.009548921, -0.0076625543, 0.009594454, 0.0065632574, 0.03746716, -0.02396337, 0.043139268, -0.015025891, 0.0023498281, 0.023377946, 0.027892217, 0.002457156, 0.027449897, -0.0269035, -0.03496935, -0.008267493, -0.017237494, 0.0011423905, -0.03489129, 0.012287406, -0.036374364, -0.0015180379, 0.002829551, -0.014765703, 0.02692952, 0.009210676, -0.032965895, -0.0040719514, 0.012183331, -0.0046638804, -0.0064559295, 0.011448299, 0.005633083, -0.042280648, -0.012443519, -0.022714464, 0.011428784, -0.024249578, -0.017692825, 0.013308646, -0.00025591988, -0.0021367988, -0.01644392, 0.0075844973, -0.00051915756, -0.018837655, 0.008449624, -0.0064266585, -0.021179352, 0.0013025692, 0.0079227425, -0.031612914, 0.0022181077, 0.044804476, 0.0191759, 0.0005305408, 0.017601758, -0.029089084, -0.019670257, 0.02371619, -0.004715918, 0.0056298305, 0.010576666, -0.0068234457, -0.010784817, -0.0072007193, 0.024509765, 0.018070098, -0.017497683, -0.027449897, -0.027892217, 0.0067909225, 0.014960844, 0.0074413936, -0.0013139524, -0.004702909, 0.016743137, 0.011903628, 0.014531533, -0.025589548, -0.012417501, 0.0063225827, -0.0054704654, -0.00078300503, -0.0125280805, -0.020060541, 0.019188909, 0.0048720315, -0.0036784161, -0.0030068045, -0.015377146, -0.004247579, -0.006124189, 0.018395334, -0.015129967, -0.016404891, -0.0043939347, 0.0046378616, -0.02305271, 0.008319531, 0.00023193374, -0.03169097, -0.00444272, -0.0016798426, -0.022272144, -0.002649045, -0.030129839, 0.00387681, -0.023221832, 0.012215855, -0.032133292, -0.030988462, 0.014180278, -0.0063258354, 0.0077796387, 0.009854643, -0.024366662, 0.022024965, 0.00027868638, -0.013646891, 0.046885986, 0.020229664, -0.024808982, -0.032029215, -0.020789068, -0.003431237, 0.0063258354, -0.0051257154, -0.0029547666, -0.015090939, 0.03627029, -0.0016733379, -0.001819694, -0.0006797427, 0.015897524, -0.024496756, 0.019423079, -0.014661627, 0.021569634, -0.014583571, -0.029401312, -0.016222758, 0.014947834, 0.008267493, 0.015988588, 0.0011862974, -0.0052623143, 0.019553173, 0.014804731, 0.011142576, 0.017315552, -0.026955537, -0.03554176, 0.019084834, 0.013672911, -0.019058814, -0.03424082, 0.028256482, 0.0071486817, -0.008482148, -0.0073112994, 0.012671185, 0.006595781, 0.007486927, 0.028906953, 0.0052265385, 0.0016456929, -0.009028545, -0.016756145, 0.018980758, 0.013386703, 0.013803005, -0.0036523973, -0.0050931918, -0.016587023, -0.011526355, -0.014674637, -0.01574141, -0.031118555, 0.009074077, 0.01346476, -0.0015619446, 0.017575739, -0.0020587423, -0.01732856, 0.02279252, 0.0013887567, -0.018993767, -0.0045630573, 0.020125588, 0.013555826, -0.0024945582, 0.017016334, 0.013334665, 0.023911333, 0.012059742, 0.027007576, -0.0148307495, 0.007213729, 0.008163417, 0.012410996, -0.013607863, -0.046469685, -0.011012482, 0.027866198, -0.0355938, 0.039990988, -0.008599233, 0.017705834, 0.052584115, 0.0029368787, -0.04496059, -0.008163417, 0.018291257, 0.02692952, -0.021491578, 0.0076365354, -0.00990668, -0.019891419, 0.020112578, -0.019566182, 0.022454275, -0.014453476, -0.028256482, 0.01800505, 0.0017432636, 0.011591402, 0.013894071, 0.012827298, 0.012580118, -0.012131293, 0.00018111565, -0.014869778, -0.04813489, -0.010043279, 0.04654774, -0.01894173, -0.018356305, -0.046209496, -0.0059908424, -0.00043093736, -0.046599776, -0.023989389, -0.003447499, -0.029557424, -0.0024685394, -0.03080633, -0.01345175, 0.008078855, -0.0059908424, 0.037571233, -0.024821991, -0.044414192, -0.034292858, -0.018993767, -0.019071825, -0.0016782164, -0.023351926], \'node_index\': 13815, \'node_id\': \'Oxygen_(14013)\', \'node_name\': \'Oxygen\'}']
CPU times: user 6.55 ms, sys: 692 μs, total: 7.24 ms
Wall time: 13.6 ms

In [15]:

%%time

vector_to_search = merged_nodes_df["desc_emb"].loc[2]


# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_nodes')

# Load the collection into memory before query
collection.load()

# Vector similarity search in Milvus
vector_to_search = normalize_vector(merged_nodes_df["desc_emb"].iloc[2]).tolist()
search_params = {"metric_type": "IP"}
results = collection.search(
    data=[vector_to_search],
    anns_field="desc_emb",
    param=search_params,
    limit=10,
    output_fields=["node_id", "node_name"]
)

results

%%time vector_to_search = merged_nodes_df["desc_emb"].loc[2] # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_nodes') # Load the collection into memory before query collection.load() # Vector similarity search in Milvus vector_to_search = normalize_vector(merged_nodes_df["desc_emb"].iloc[2]).tolist() search_params = {"metric_type": "IP"} results = collection.search( data=[vector_to_search], anns_field="desc_emb", param=search_params, limit=10, output_fields=["node_id", "node_name"] ) results

CPU times: user 140 ms, sys: 39.1 ms, total: 179 ms
Wall time: 521 ms

Out[15]:

data: [[{'node_index': 11504, 'distance': 1.0000001192092896, 'entity': {'node_name': 'SSTR2', 'node_id': 'SSTR2_(11630)'}}, {'node_index': 28075, 'distance': 0.9640881419181824, 'entity': {'node_name': 'SSTR4', 'node_id': 'SSTR4_(34965)'}}, {'node_index': 13207, 'distance': 0.9590871334075928, 'entity': {'node_name': 'SSTR1', 'node_id': 'SSTR1_(13376)'}}, {'node_index': 10873, 'distance': 0.9556770920753479, 'entity': {'node_name': 'SSTR3', 'node_id': 'SSTR3_(10983)'}}, {'node_index': 11786, 'distance': 0.9445285797119141, 'entity': {'node_name': 'SSTR5', 'node_id': 'SSTR5_(11920)'}}, {'node_index': 12024, 'distance': 0.9246837496757507, 'entity': {'node_name': 'SST', 'node_id': 'SST_(12165)'}}, {'node_index': 15309, 'distance': 0.9021860361099243, 'entity': {'node_name': 'Somatostatin', 'node_id': 'Somatostatin_(15618)'}}, {'node_index': 10048, 'distance': 0.8932695388793945, 'entity': {'node_name': 'SCTR', 'node_id': 'SCTR_(10137)'}}, {'node_index': 78455, 'distance': 0.891120433807373, 'entity': {'node_name': 'type 2 somatostatin receptor binding', 'node_id': 'type 2 somatostatin receptor binding_(120756)'}}, {'node_index': 11639, 'distance': 0.8880254030227661, 'entity': {'node_name': 'SSR2', 'node_id': 'SSR2_(11769)'}}]]

In [16]:

merged_nodes_df.loc[2]

merged_nodes_df.loc[2]

Out[16]:

	node_index	primekg_node_index	node_id	node_name	node_type	desc	feat	desc_emb	feat_emb
2	11504	11630	SSTR2_(11630)	SSTR2	gene/protein	SSTR2 belongs to gene/protein node. SSTR2 is s...	MDMADEPLNGSHTWLSIPFDLNGSVVSTNTSNQTEPYYDLTSNAVL...	[-0.036013797, 0.002569047, -0.0049933116, -0....	[0.10192450135946274, -0.04799338057637215, 0....

Building Node Collection (Node Type-specific Embedding)¶

Note that nodes information of the PrimeKG data is different for each node type, we are going to build a separate collection for each node type.

We will use the node type as the collection name.

In [15]:

%%time

# Loop over group enrichment nodes by node_type
for node_type, nodes_df in tqdm(merged_nodes_df.groupby('node_type')):
    print(f"Processing node type: {node_type}")

    # Milvus collection name for this node_type
    node_coll_name = f"{milvus_database}_nodes_{node_type.replace('/', '_')}"

    # Define collection schema
    desc_emb_dim = len(nodes_df.iloc[0]['desc_emb'].to_arrow().to_pylist()[0])
    feat_emb_dim = len(nodes_df.iloc[0]['feat_emb'].to_arrow().to_pylist()[0])
    node_fields = [
        FieldSchema(name="node_index", dtype=DataType.INT64, is_primary=True, auto_id=False),
        FieldSchema(name="node_id", dtype=DataType.VARCHAR, max_length=1024),
        FieldSchema(name="node_name", dtype=DataType.VARCHAR, max_length=1024,
                    enable_analyzer=True, enable_match=True),
        FieldSchema(name="node_type", dtype=DataType.VARCHAR, max_length=1024,
                    enable_analyzer=True, enable_match=True),
        FieldSchema(name="desc", dtype=DataType.VARCHAR, max_length=40960,
                    enable_analyzer=True, enable_match=True),
        FieldSchema(name="desc_emb", dtype=DataType.FLOAT_VECTOR, dim=desc_emb_dim),        
        FieldSchema(name="feat", dtype=DataType.VARCHAR, max_length=40960,
                    enable_analyzer=True, enable_match=True),
        FieldSchema(name="feat_emb", dtype=DataType.FLOAT_VECTOR, dim=feat_emb_dim),
    ]
    schema = CollectionSchema(fields=node_fields, description=f"schema for collection {node_coll_name}")

    # Create collection if not exists
    if not utility.has_collection(node_coll_name):
        collection = Collection(name=node_coll_name, schema=schema)
    else:
        collection = Collection(name=node_coll_name)

    # Create index for node_index field (scalar)
    collection.create_index(
        field_name="node_index",
        index_params={"index_type": "STL_SORT"},
        index_name="node_index_index"
    )
    # Create index for node_name, node_type, desc fields (inverted)
    collection.create_index(
        field_name="node_name",
        index_params={"index_type": "INVERTED"},
        index_name="node_name_index"
    )
    collection.create_index(
        field_name="node_type",
        index_params={"index_type": "INVERTED"},
        index_name="node_type_index"
    )
    collection.create_index(
        field_name="desc",
        index_params={"index_type": "INVERTED"},
        index_name="desc_index"
    )
    collection.create_index(
        field_name="desc_emb",
        index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX
        index_name="desc_emb_index"
    )
    # Create index for feat_emb (vector)
    collection.create_index(
        field_name="feat_emb",
        index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX
        index_name="feat_emb_index"
    )

    # Prepare data for insertion
    # Normalize the embeddings
    graph_desc_emb_cp = cp.asarray(nodes_df["desc_emb"].list.leaves).astype(cp.float32).reshape(nodes_df.shape[0], -1)
    graph_desc_emb_norm = normalize_matrix(graph_desc_emb_cp, axis=1)
    graph_feat_emb_cp = cp.asarray(nodes_df["feat_emb"].list.leaves).astype(cp.float32).reshape(nodes_df.shape[0], -1)
    graph_feat_emb_norm = normalize_matrix(graph_feat_emb_cp, axis=1)
    # Columns must be lists of values in order matching schema fields
    data = [
        nodes_df["node_index"].to_arrow().to_pylist(),
        nodes_df["node_id"].to_arrow().to_pylist(),
        nodes_df["node_name"].to_arrow().to_pylist(),
        nodes_df["node_type"].to_arrow().to_pylist(),
        nodes_df["desc"].to_arrow().to_pylist(),
        graph_desc_emb_norm.tolist(), # Use normalized embeddings
        nodes_df["feat"].to_arrow().to_pylist(),
        graph_feat_emb_norm.tolist(), # Use normalized embeddings
    ]

    # Batch insert data in chunks
    batch_size = 500
    total_rows = len(data[0])
    for i in tqdm(range(0, total_rows, batch_size)):
        batch = [col[i:i + batch_size] for col in data]
        collection.insert(batch)

    # Flush the collection to ensure data is persisted
    collection.flush()

    # Print collection stats (number of entities and segment info)
    stats = collection.num_entities
    print(f"Collection {node_coll_name} stats:")
    print(stats)

%%time # Loop over group enrichment nodes by node_type for node_type, nodes_df in tqdm(merged_nodes_df.groupby('node_type')): print(f"Processing node type: {node_type}") # Milvus collection name for this node_type node_coll_name = f"{milvus_database}_nodes_{node_type.replace('/', '_')}" # Define collection schema desc_emb_dim = len(nodes_df.iloc[0]['desc_emb'].to_arrow().to_pylist()[0]) feat_emb_dim = len(nodes_df.iloc[0]['feat_emb'].to_arrow().to_pylist()[0]) node_fields = [ FieldSchema(name="node_index", dtype=DataType.INT64, is_primary=True, auto_id=False), FieldSchema(name="node_id", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="node_name", dtype=DataType.VARCHAR, max_length=1024, enable_analyzer=True, enable_match=True), FieldSchema(name="node_type", dtype=DataType.VARCHAR, max_length=1024, enable_analyzer=True, enable_match=True), FieldSchema(name="desc", dtype=DataType.VARCHAR, max_length=40960, enable_analyzer=True, enable_match=True), FieldSchema(name="desc_emb", dtype=DataType.FLOAT_VECTOR, dim=desc_emb_dim), FieldSchema(name="feat", dtype=DataType.VARCHAR, max_length=40960, enable_analyzer=True, enable_match=True), FieldSchema(name="feat_emb", dtype=DataType.FLOAT_VECTOR, dim=feat_emb_dim), ] schema = CollectionSchema(fields=node_fields, description=f"schema for collection {node_coll_name}") # Create collection if not exists if not utility.has_collection(node_coll_name): collection = Collection(name=node_coll_name, schema=schema) else: collection = Collection(name=node_coll_name) # Create index for node_index field (scalar) collection.create_index( field_name="node_index", index_params={"index_type": "STL_SORT"}, index_name="node_index_index" ) # Create index for node_name, node_type, desc fields (inverted) collection.create_index( field_name="node_name", index_params={"index_type": "INVERTED"}, index_name="node_name_index" ) collection.create_index( field_name="node_type", index_params={"index_type": "INVERTED"}, index_name="node_type_index" ) collection.create_index( field_name="desc", index_params={"index_type": "INVERTED"}, index_name="desc_index" ) collection.create_index( field_name="desc_emb", index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX index_name="desc_emb_index" ) # Create index for feat_emb (vector) collection.create_index( field_name="feat_emb", index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, # AUTOINDEX index_name="feat_emb_index" ) # Prepare data for insertion # Normalize the embeddings graph_desc_emb_cp = cp.asarray(nodes_df["desc_emb"].list.leaves).astype(cp.float32).reshape(nodes_df.shape[0], -1) graph_desc_emb_norm = normalize_matrix(graph_desc_emb_cp, axis=1) graph_feat_emb_cp = cp.asarray(nodes_df["feat_emb"].list.leaves).astype(cp.float32).reshape(nodes_df.shape[0], -1) graph_feat_emb_norm = normalize_matrix(graph_feat_emb_cp, axis=1) # Columns must be lists of values in order matching schema fields data = [ nodes_df["node_index"].to_arrow().to_pylist(), nodes_df["node_id"].to_arrow().to_pylist(), nodes_df["node_name"].to_arrow().to_pylist(), nodes_df["node_type"].to_arrow().to_pylist(), nodes_df["desc"].to_arrow().to_pylist(), graph_desc_emb_norm.tolist(), # Use normalized embeddings nodes_df["feat"].to_arrow().to_pylist(), graph_feat_emb_norm.tolist(), # Use normalized embeddings ] # Batch insert data in chunks batch_size = 500 total_rows = len(data[0]) for i in tqdm(range(0, total_rows, batch_size)): batch = [col[i:i + batch_size] for col in data] collection.insert(batch) # Flush the collection to ensure data is persisted collection.flush() # Print collection stats (number of entities and segment info) stats = collection.num_entities print(f"Collection {node_coll_name} stats:") print(stats)

  0%|          | 0/6 [00:00<?, ?it/s]

Processing node type: biological_process

100%|██████████| 55/55 [00:48<00:00,  1.12it/s]
 17%|█▋        | 1/6 [00:59<04:57, 59.56s/it]

Collection t2kg_primekg_nodes_biological_process stats:
27409
Processing node type: cellular_component

100%|██████████| 9/9 [00:07<00:00,  1.24it/s]
 33%|███▎      | 2/6 [01:14<02:12, 33.03s/it]

Collection t2kg_primekg_nodes_cellular_component stats:
4011
Processing node type: disease

100%|██████████| 35/35 [00:30<00:00,  1.15it/s]
 50%|█████     | 3/6 [01:51<01:45, 35.14s/it]

Collection t2kg_primekg_nodes_disease stats:
17054
Processing node type: drug

100%|██████████| 14/14 [00:11<00:00,  1.18it/s]
 67%|██████▋   | 4/6 [02:08<00:56, 28.09s/it]

Collection t2kg_primekg_nodes_drug stats:
6759
Processing node type: gene/protein

100%|██████████| 38/38 [00:41<00:00,  1.09s/it]
 83%|████████▎ | 5/6 [02:58<00:35, 35.68s/it]

Collection t2kg_primekg_nodes_gene_protein stats:
18797
Processing node type: molecular_function

100%|██████████| 22/22 [00:19<00:00,  1.14it/s]
100%|██████████| 6/6 [03:24<00:00, 34.10s/it]

Collection t2kg_primekg_nodes_molecular_function stats:
10951
CPU times: user 1min 5s, sys: 8.56 s, total: 1min 14s
Wall time: 3min 24s

In [16]:

# List all collections
for coll in utility.list_collections():
    print(f"Collection: {coll}")

    # Load the collection to get stats
    collection = Collection(name=coll)
    print(collection.num_entities)

# List all collections for coll in utility.list_collections(): print(f"Collection: {coll}") # Load the collection to get stats collection = Collection(name=coll) print(collection.num_entities)

Collection: t2kg_primekg_nodes_disease
17054
Collection: t2kg_primekg_nodes_drug
6759
Collection: t2kg_primekg_nodes_gene_protein
18797
Collection: t2kg_primekg_nodes_molecular_function
10951
Collection: t2kg_primekg_nodes
84981
Collection: t2kg_primekg_nodes_biological_process
27409
Collection: t2kg_primekg_nodes_cellular_component
4011

In [12]:

%%time

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_nodes_gene_protein')

# Load the collection into memory before query
collection.load()

# Vector similarity search in Milvus
vector_to_search = normalize_vector(merged_nodes_df["feat_emb"].iloc[2]).tolist()
search_params = {"metric_type": "IP"}
results = collection.search(
    data=[vector_to_search],
    anns_field="feat_emb",
    param=search_params,
    limit=10,
    output_fields=["node_id", "node_name"]
)

results

%%time # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_nodes_gene_protein') # Load the collection into memory before query collection.load() # Vector similarity search in Milvus vector_to_search = normalize_vector(merged_nodes_df["feat_emb"].iloc[2]).tolist() search_params = {"metric_type": "IP"} results = collection.search( data=[vector_to_search], anns_field="feat_emb", param=search_params, limit=10, output_fields=["node_id", "node_name"] ) results

CPU times: user 13.9 ms, sys: 366 μs, total: 14.2 ms
Wall time: 22.9 ms

Out[12]:

data: [[{'node_index': 7072, 'distance': 0.9999998807907104, 'entity': {'node_id': 'SOX14_(7119)', 'node_name': 'SOX14'}}, {'node_index': 460, 'distance': 0.9957526922225952, 'entity': {'node_id': 'SOX2_(460)', 'node_name': 'SOX2'}}, {'node_index': 13546, 'distance': 0.9949035048484802, 'entity': {'node_id': 'SOX7_(13730)', 'node_name': 'SOX7'}}, {'node_index': 49952, 'distance': 0.9946167469024658, 'entity': {'node_id': 'SOX21_(57709)', 'node_name': 'SOX21'}}, {'node_index': 6972, 'distance': 0.9945659041404724, 'entity': {'node_id': 'SOX9_(7019)', 'node_name': 'SOX9'}}, {'node_index': 9392, 'distance': 0.9940434694290161, 'entity': {'node_id': 'LEF1_(9476)', 'node_name': 'LEF1'}}, {'node_index': 1678, 'distance': 0.9940416216850281, 'entity': {'node_id': 'SOX10_(1682)', 'node_name': 'SOX10'}}, {'node_index': 11064, 'distance': 0.9938754439353943, 'entity': {'node_id': 'TOX2_(11177)', 'node_name': 'TOX2'}}, {'node_index': 6659, 'distance': 0.993757426738739, 'entity': {'node_id': 'TOX_(6704)', 'node_name': 'TOX'}}, {'node_index': 5090, 'distance': 0.9937493205070496, 'entity': {'node_id': 'SOX30_(5120)', 'node_name': 'SOX30'}}]]

In [13]:

# Check the ground truth for the search
merged_nodes_df.loc[2]

# Check the ground truth for the search merged_nodes_df.loc[2]

Out[13]:

	node_index	primekg_node_index	node_id	node_name	node_type	desc	feat	desc_emb	feat_emb
2	7072	7119	SOX14_(7119)	SOX14	gene/protein	SOX14 belongs to gene/protein node. SOX14 is S...	MSKPSDHIKRPMNAFMVWSRGQRRKMAQENPKMHNSEISKRLGAEW...	[-0.02060385, -0.020443613, -0.017091982, -0.0...	[0.049131330102682114, 0.014837502501904964, 0...

In [14]:

# Get node indices from the results
[n['node_index'] for n in results[0]]

# Get node indices from the results [n['node_index'] for n in results[0]]

Out[14]:

[7072, 460, 13546, 49952, 6972, 9392, 1678, 11064, 6659, 5090]

In [15]:

# Get the cosine similarity scores
[n['distance'] for n in results[0]]

# Get the cosine similarity scores [n['distance'] for n in results[0]]

Out[15]:

[0.9999998807907104,
 0.9957526922225952,
 0.9949035048484802,
 0.9946167469024658,
 0.9945659041404724,
 0.9940434694290161,
 0.9940416216850281,
 0.9938754439353943,
 0.993757426738739,
 0.9937493205070496]

Building Edge Collection¶

Subsquently, we are also building the edges collection in Milvus.

Note that the edges information of PrimeKG has massive records, so once again we are chunking the data to avoid memory issues.

In [17]:

%%time

# Define collection name
edge_coll_name = f"{milvus_database}_edges"

# Define schema
edge_fields = [
    FieldSchema(name="triplet_index", dtype=DataType.INT64, is_primary=True, auto_id=False),
    FieldSchema(name="head_id", dtype=DataType.VARCHAR, max_length=1024),
    FieldSchema(name="head_index", dtype=DataType.INT64),
    FieldSchema(name="tail_id", dtype=DataType.VARCHAR, max_length=1024),
    FieldSchema(name="tail_index", dtype=DataType.INT64),
    FieldSchema(name="edge_type", dtype=DataType.VARCHAR, max_length=1024),
    FieldSchema(name="display_relation", dtype=DataType.VARCHAR, max_length=1024),
    FieldSchema(name="feat", dtype=DataType.VARCHAR, max_length=40960),
    FieldSchema(name="feat_emb", dtype=DataType.FLOAT_VECTOR, dim=1536),
]
edge_schema = CollectionSchema(fields=edge_fields, description="Schema for edges collection")

# Create collection if not exists
if not utility.has_collection(edge_coll_name):
    collection = Collection(name=edge_coll_name, schema=edge_schema)
else:
    collection = Collection(name=edge_coll_name)

# Create indexes
collection.create_index(field_name="triplet_index", index_params={"index_type": "STL_SORT"}, index_name="triplet_index_index")
collection.create_index(field_name="head_index", index_params={"index_type": "STL_SORT"}, index_name="head_index_index")
collection.create_index(field_name="tail_index", index_params={"index_type": "STL_SORT"}, index_name="tail_index_index")
collection.create_index(field_name="feat_emb", index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, index_name="feat_emb_index") # AUTOINDEX

# Iterate over chunked edges embedding df
for edges_df in tqdm(edges_embedding_df):
    # Merge enrichment with embedding
    merged_edges_df = edges_enrichment_df.merge(
        edges_df[["triplet_index", "edge_emb"]],
        on="triplet_index",
        how="inner"
    )

    # Prepare data fields in column-wise format
    # Normalize the embeddings
    edges_edge_emb_cp = cp.asarray(merged_edges_df["edge_emb"].list.leaves).astype(cp.float32).reshape(merged_edges_df.shape[0], -1)
    edges_edge_emb_norm = normalize_matrix(edges_edge_emb_cp, axis=1)
    data = [
        merged_edges_df["triplet_index"].to_arrow().to_pylist(),
        merged_edges_df["head_id"].to_arrow().to_pylist(),
        merged_edges_df["head_index"].to_arrow().to_pylist(),
        merged_edges_df["tail_id"].to_arrow().to_pylist(),
        merged_edges_df["tail_index"].to_arrow().to_pylist(),
        merged_edges_df["edge_type_str"].to_arrow().to_pylist(),
        merged_edges_df["display_relation"].to_arrow().to_pylist(),
        merged_edges_df["feat"].to_arrow().to_pylist(),
        edges_edge_emb_norm.tolist(), # Use normalized embeddings
    ]

    # Insert in chunks
    batch_size = 500
    for i in tqdm(range(0, len(data[0]), batch_size)):
        batch_data = [d[i:i+batch_size] for d in data]
        collection.insert(batch_data)

    # Flush to ensure persistence
    collection.flush()

    # Print collection stats
    print(collection.num_entities)
    
    time.sleep(5)  # Sleep to avoid overwhelming the server

%%time # Define collection name edge_coll_name = f"{milvus_database}_edges" # Define schema edge_fields = [ FieldSchema(name="triplet_index", dtype=DataType.INT64, is_primary=True, auto_id=False), FieldSchema(name="head_id", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="head_index", dtype=DataType.INT64), FieldSchema(name="tail_id", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="tail_index", dtype=DataType.INT64), FieldSchema(name="edge_type", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="display_relation", dtype=DataType.VARCHAR, max_length=1024), FieldSchema(name="feat", dtype=DataType.VARCHAR, max_length=40960), FieldSchema(name="feat_emb", dtype=DataType.FLOAT_VECTOR, dim=1536), ] edge_schema = CollectionSchema(fields=edge_fields, description="Schema for edges collection") # Create collection if not exists if not utility.has_collection(edge_coll_name): collection = Collection(name=edge_coll_name, schema=edge_schema) else: collection = Collection(name=edge_coll_name) # Create indexes collection.create_index(field_name="triplet_index", index_params={"index_type": "STL_SORT"}, index_name="triplet_index_index") collection.create_index(field_name="head_index", index_params={"index_type": "STL_SORT"}, index_name="head_index_index") collection.create_index(field_name="tail_index", index_params={"index_type": "STL_SORT"}, index_name="tail_index_index") collection.create_index(field_name="feat_emb", index_params={"index_type": "GPU_CAGRA", "metric_type": "IP"}, index_name="feat_emb_index") # AUTOINDEX # Iterate over chunked edges embedding df for edges_df in tqdm(edges_embedding_df): # Merge enrichment with embedding merged_edges_df = edges_enrichment_df.merge( edges_df[["triplet_index", "edge_emb"]], on="triplet_index", how="inner" ) # Prepare data fields in column-wise format # Normalize the embeddings edges_edge_emb_cp = cp.asarray(merged_edges_df["edge_emb"].list.leaves).astype(cp.float32).reshape(merged_edges_df.shape[0], -1) edges_edge_emb_norm = normalize_matrix(edges_edge_emb_cp, axis=1) data = [ merged_edges_df["triplet_index"].to_arrow().to_pylist(), merged_edges_df["head_id"].to_arrow().to_pylist(), merged_edges_df["head_index"].to_arrow().to_pylist(), merged_edges_df["tail_id"].to_arrow().to_pylist(), merged_edges_df["tail_index"].to_arrow().to_pylist(), merged_edges_df["edge_type_str"].to_arrow().to_pylist(), merged_edges_df["display_relation"].to_arrow().to_pylist(), merged_edges_df["feat"].to_arrow().to_pylist(), edges_edge_emb_norm.tolist(), # Use normalized embeddings ] # Insert in chunks batch_size = 500 for i in tqdm(range(0, len(data[0]), batch_size)): batch_data = [d[i:i+batch_size] for d in data] collection.insert(batch_data) # Flush to ensure persistence collection.flush() # Print collection stats print(collection.num_entities) time.sleep(5) # Sleep to avoid overwhelming the server

100%|██████████| 500/500 [05:31<00:00,  1.51it/s]

250000

100%|██████████| 500/500 [05:34<00:00,  1.49it/s]

500000

100%|██████████| 500/500 [05:36<00:00,  1.49it/s]

750000

100%|██████████| 500/500 [05:34<00:00,  1.49it/s]

1000000

100%|██████████| 500/500 [05:33<00:00,  1.50it/s]

1250000

100%|██████████| 410/410 [04:34<00:00,  1.49it/s]

1454610

100%|██████████| 500/500 [05:36<00:00,  1.49it/s]

1704610

100%|██████████| 500/500 [05:34<00:00,  1.50it/s]

1954610

100%|██████████| 500/500 [05:33<00:00,  1.50it/s]

2204610

100%|██████████| 500/500 [05:34<00:00,  1.50it/s]

2454610

100%|██████████| 500/500 [05:36<00:00,  1.49it/s]]

2704610

100%|██████████| 500/500 [05:31<00:00,  1.51it/s]]

2954610

100%|██████████| 500/500 [05:36<00:00,  1.49it/s]]

3204610

100%|██████████| 500/500 [05:33<00:00,  1.50it/s]]

3454610

100%|██████████| 500/500 [05:34<00:00,  1.49it/s]]

3704610

100%|██████████| 400/400 [04:27<00:00,  1.50it/s]]

3904610

100%|██████████| 16/16 [1:35:30<00:00, 358.18s/it]

CPU times: user 23min 24s, sys: 2min 56s, total: 26min 20s
Wall time: 1h 35min 33s

In [6]:

# List all collections
for coll in utility.list_collections():
    print(f"Collection: {coll}")

    # Load the collection to get stats
    collection = Collection(name=coll)
    print(collection.num_entities)

# List all collections for coll in utility.list_collections(): print(f"Collection: {coll}") # Load the collection to get stats collection = Collection(name=coll) print(collection.num_entities)

Collection: t2kg_primekg_nodes_molecular_function
10951
Collection: t2kg_primekg_edges
3904610
Collection: t2kg_primekg_nodes
84981
Collection: t2kg_primekg_nodes_biological_process
27409
Collection: t2kg_primekg_nodes_cellular_component
4011
Collection: t2kg_primekg_nodes_disease
17054
Collection: t2kg_primekg_nodes_drug
6759
Collection: t2kg_primekg_nodes_gene_protein
18797

In [7]:

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_edges')

# Load the collection into memory before query
collection.load()

# Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_edges') # Load the collection into memory before query collection.load()

In [8]:

%%time

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_edges')

# Load the collection into memory before query
collection.load()

# Query by expr on triplet_index
expr = "triplet_index == 0"
output_fields = ["triplet_index", "head_id", "tail_id", "edge_type", "feat", "feat_emb"]

results = collection.query(expr, output_fields=output_fields)
results

%%time # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_edges') # Load the collection into memory before query collection.load() # Query by expr on triplet_index expr = "triplet_index == 0" output_fields = ["triplet_index", "head_id", "tail_id", "edge_type", "feat", "feat_emb"] results = collection.query(expr, output_fields=output_fields) results

CPU times: user 5 ms, sys: 0 ns, total: 5 ms
Wall time: 564 ms

Out[8]:

data: ["{'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with KIF15 (gene/protein).', 'feat_emb': [-0.01934238, 0.0011752498, 0.004431808, -0.033904973, -0.01556057, 0.011148459, 0.0011645806, 0.028704984, 0.012691384, -0.005199988, 0.020970657, 0.01423431, 0.009421695, 0.0033714569, -0.0027230997, -0.0106954295, 0.015586833, -0.015100975, 0.01833127, 0.014812087, -0.008581293, 0.019014098, 0.005584078, -0.036977693, 0.010511592, 0.034535274, 0.010833308, -0.026236303, -0.025107013, -0.037502944, 0.027575694, -0.019513085, -0.001110414, 0.015416126, 0.0045368583, -0.0110171465, 0.009841897, -0.021863585, 0.016400972, -0.0021190608, -0.0024818124, -0.0077540223, -0.006516399, 0.0013426737, -0.0038474659, -0.0059189256, 0.0074257404, -0.022231262, -0.017438343, 0.008292405, -0.014102997, 0.002123985, -0.012579769, -0.020051468, 0.0068545295, -0.0032368612, 0.0050850892, 0.00063440506, -0.028127206, -0.004412111, 0.0011465251, 0.0029463316, -0.008568162, -0.0091984635, 0.016532285, 0.006933317, 0.0019582026, 0.023124188, 0.015219157, -0.0047896355, 0.04125849, 0.033012044, -0.0026837059, -0.004812615, 0.013761584, -0.0022700704, -0.007970689, 0.0009544801, -0.02783832, 0.024411054, 0.019197935, -0.0050653922, 0.018580765, 0.0091984635, 0.015258551, -0.03091104, -0.017293898, 0.029729223, 0.020668639, -0.009828765, 0.02917771, 0.020629246, 0.003824486, 0.004855292, -0.008141396, -0.0114636095, -0.0006499985, 0.01269795, 0.011082803, -0.021797929, -0.026919128, 0.029414073, -0.014680774, -0.0013032798, -0.018055514, -0.0077737193, -0.010012603, -0.0108595705, 0.0109383585, -0.013879766, -0.012967141, 0.03758173, 0.012803, -0.02343934, 0.008587859, -0.00035433946, 0.031357504, 0.013800978, -0.012356536, -0.0041133743, 0.031173665, 0.0021912828, 0.009579271, -0.013288858, 0.026354484, 0.03726658, -0.021023184, -0.024319135, -0.009585836, -0.0054855933, 0.018042382, 0.0070514986, 0.011227246, -0.0059583196, -0.017792888, -0.0072944276, -0.018777734, -0.016466629, -0.004740393, -0.024437318, 0.005597209, 0.022966614, -0.028915085, 0.0084302835, -0.013866634, 0.012691384, 0.010065128, -0.003604537, -0.020392882, 0.0019417884, 0.004671454, -0.0021666617, -0.0116671445, 0.013439868, 0.013498958, 0.004192162, -0.0026295392, 0.00427095, -0.002288126, -0.00012382389, -0.013590878, -2.7416681e-05, 0.0012901485, 0.012468153, 0.03041205, 0.030674675, -0.0035782745, -0.011128762, 0.005787613, -0.02447671, 0.0031465837, 0.015337338, -0.059300873, 0.01402421, 0.022717118, 0.00039968343, 0.021351466, 0.011187852, -0.027365593, -0.018278746, -0.011798457, 0.00046410877, 0.016913092, -0.0022470907, 0.005180291, -0.0042020106, 0.027759532, -0.023019139, -0.004392414, -0.015993902, 0.014549461, 0.014601986, 0.012126739, -0.0051277657, -0.645849, -0.040785763, 0.007557053, -0.0134464335, 0.027391855, 0.033038307, 0.005213119, 0.014851481, 0.015100975, 0.0068348325, -0.0017005011, 0.03288073, -0.014995925, -0.017228242, -0.015796933, -0.03277568, -0.011115631, -0.047955442, 0.00282815, -0.014772693, -0.013906028, -0.012954011, 0.01402421, 0.033012044, 0.016781779, -0.0012606031, 0.05005645, -0.0058467034, -0.00011448837, 0.016729254, -0.019224199, 0.033773657, 0.0032614823, 0.011647448, 0.039288796, -0.0060469555, 0.013564615, 0.01914541, 0.010879268, 0.03471911, -0.032197904, -6.422018e-05, 0.0131247165, 0.013551484, -0.021049445, 0.016939355, 0.008994929, 0.00028252776, 0.022612069, -0.0046222117, -0.024621155, 0.011864114, -0.011686841, 0.009231292, -0.0049209483, 0.0058926633, 0.026236303, -0.03490295, -0.0040345867, 0.002691913, 0.008876747, -0.0076227095, -0.002636105, -0.01161462, 0.014247442, 0.010183309, -0.004300495, -0.0039689303, 6.873406e-05, -0.0074191745, 0.029151447, 0.011864114, -0.004796201, 0.006736348, 0.004684585, 0.03724032, 0.027391855, 0.02712923, -0.0091196755, 0.018055514, -0.024949437, 0.013170676, -0.021154497, 0.012074214, 0.0144969355, 0.00885705, -0.02642014, 0.0131247165, -0.008627253, -0.025356507, 0.015429257, -0.029702961, -0.012356536, -0.018685816, -0.030333262, -0.006332561, -0.03666254, -0.005961602, 0.036504965, -0.028074682, 0.0049570594, 0.0034141336, -0.019184804, 0.0030530232, 0.020747427, -0.007911597, -0.0249757, -0.007872204, 0.01956561, 0.001959844, -0.0037818095, -0.01233684, -0.0030530232, -0.017727232, -0.018501977, -0.032197904, 0.05783017, -0.001848228, -0.0007767974, -0.010682299, 0.0006011665, 0.007845941, 0.00056546583, 0.002544186, 0.02682721, 0.028100945, 0.00463206, -0.023767622, -0.007845941, -0.013997947, 0.017018143, -0.004251253, 0.02580297, -0.0023701964, 0.012304012, 0.014076735, 0.018055514, 0.010577248, 0.022743382, -0.034430224, 4.362561e-05, -0.0023981004, 0.013374211, -0.014720168, -0.0008256294, -0.025658526, -0.029545385, -0.01361714, 0.033773657, -0.0016873698, 0.01802925, -0.011936336, 0.0009052378, 0.022047423, 0.0155080445, -0.006388369, -0.003156432, 0.01607269, -0.009559574, -0.030884776, 0.012566637, 0.0014148956, -0.02519893, -0.019014098, -0.019158542, -0.023478733, 0.022966614, 0.005945188, -0.007537356, -0.004612363, 0.0048421603, -0.026144383, -0.022874694, 0.012284315, -0.0033386287, 0.010097956, 0.034141336, -0.0070449333, -0.0014017645, -0.00988129, -0.0025326961, -0.017700968, -0.027286805, 0.010459066, 0.03337972, 0.0015183046, 0.013590878, -0.0014797314, 0.0025770143, 0.033327196, -0.026577717, 0.013656534, -0.0043825656, 0.009717149, -0.0029217103, -0.015836326, 0.01956561, 0.0030595888, -0.010688864, 0.014286836, 0.028311046, -0.026971655, 0.004996453, -0.046852417, 0.009592402, -0.0076555377, -0.021233283, -0.022257524, -0.0052590785, -0.007839376, 0.009907553, 0.006368672, -0.0028379986, -0.010229269, 0.0033369872, -0.014890875, -0.014680774, 0.0072222054, -0.008443414, -0.0033205731, 0.017740363, 0.01054442, 0.02263833, 0.008981797, -0.011299469, -0.00988129, 0.02345247, 0.010051996, 0.012448456, -0.010951489, -0.00691362, -0.0049734735, 0.021561567, 0.016361577, -0.00030407126, -0.009283816, 0.002435853, -0.011345428, 0.0253959, 0.013787847, 0.00011571943, 0.009283816, 0.012546941, -0.023334289, 0.005561098, -0.011141893, 0.022507017, 0.0024736053, -0.025330244, -0.0012064365, 0.005163877, 0.025422163, -0.007891901, 0.010025734, 0.00655251, -0.011299469, 0.0035290322, 0.030438313, 0.042440306, 0.011325731, -0.0068676607, -0.011292903, 0.0134858275, 0.005193422, 0.00040809563, 0.011647448, -0.0144969355, -0.0057121078, -0.03183023, -0.011200984, -0.0053509977, 0.021207022, 0.0018039099, -0.02416156, 0.014785824, 0.007176246, 0.022270655, 0.0109383585, -0.00813483, 0.0015297944, -0.027155492, -0.033248406, 0.015100975, 0.033222146, -0.015074712, -0.0057547847, -0.033117093, 0.0155474385, -0.0063391267, 0.0144969355, -0.00668054, 0.016873699, 0.03532315, 0.0018350967, -0.011325731, -0.025461556, 0.0042348388, -0.00998634, 0.021180758, -0.0039098393, -0.019526217, 0.005393674, -0.034062546, -3.1391974e-05, 0.06812509, 0.015429257, -0.008594424, 0.003181053, -0.0033140075, -0.0027214582, 0.029466597, -0.00077023177, 0.008410586, 0.0011867397, 0.001102207, -0.0058696833, -0.016847435, 0.0029381246, 0.0020862324, -0.0006602573, -0.016046427, -0.036268603, -0.012146437, 0.00030571266, 0.013334817, 0.031672653, 0.008463112, 0.005413371, 0.00060444936, 0.0074060434, -0.009756543, -0.019211067, 0.011634316, -0.010314623, 0.0026837059, -0.019486824, 0.0057744817, 0.0049439277, 0.0056595826, 0.006880792, 0.0018958289, -0.04737767, -0.011988861, 0.007327256, -0.013295423, 0.0014009437, 0.01864642, 0.023255501, -0.007287862, -0.003936102, 0.040733237, 0.032302953, 0.030753464, -0.012809566, 0.0071302867, -0.0013574463, 0.007281296, -0.014707036, -0.02519893, -0.024529235, 0.0032319368, 0.00307272, 0.023084795, 0.0055315527, 0.0021994899, 0.012139871, 0.012107043, -0.012264618, 0.019132279, 0.0020369901, 0.005025998, -0.0033369872, -0.005987865, -0.013945422, 0.010682299, 0.009093413, -0.014877743, -0.022546412, 0.0108989645, -0.01842319, 0.008929272, -0.016702991, 0.009513614, -0.022086818, -0.0012450097, -0.019893894, -0.0075898813, -0.001776006, 0.014037341, -0.0063391267, 0.0099666435, -0.03471911, -0.012481284, -0.018922178, -0.00014516222, -0.020484801, -0.0069070547, -0.0065065506, -0.017779756, 0.0061946823, 0.005196705, -0.0032155227, 0.021666616, 0.0013993023, -0.018790865, 0.0015962715, 0.02366257, -0.009139373, 0.0012663481, 0.033721134, -0.0027821905, 0.001995955, 0.019316116, -0.0077343252, -0.013223202, 0.027286805, 0.006240642, -0.018685816, -0.015311075, 0.00088554085, 0.02087874, 0.012061083, -0.019027228, 0.0029315588, 0.019276723, -0.033117093, -0.0037653954, -0.022703988, 0.008167658, -0.000718117, -0.011240378, 0.0038999908, -0.036610015, -0.012901485, 0.0051212003, 2.8852915e-05, 0.005025998, -0.0025294132, 0.01013735, 0.03112114, -0.0077999816, 0.007228771, 7.309405e-05, -0.02078682, -0.023097927, 0.0075701843, 0.018856522, -0.00282815, -0.03041205, 0.014523199, -0.01351209, -0.028731246, -0.021338334, 0.014405017, -0.038422134, 0.030175688, 0.012737344, -0.019736318, -0.0202747, -0.005025998, -0.04104839, -0.013052495, -0.025041357, -0.0053411494, 0.0033304216, -0.023465602, 0.018685816, -0.005790896, 0.018593896, -0.024686811, -0.03264437, 0.01432623, 0.023583783, 0.0043530203, -0.0035552948, 0.0074454374, -0.020314094, -0.0074979626, 0.0014157164, -0.037109006, -0.0075110937, -0.0050752405, 0.023938328, 0.020235306, 0.0300969, 0.008686343, 0.013387343, 0.028284783, 0.011227246, -0.011529266, 0.018147433, 0.002424363, -0.044173636, 0.024292873, 0.021036314, -0.022979744, 0.026945392, 0.0043234746, 0.0029184276, 0.026603978, -0.007524225, -0.018515108, -0.0040148897, -0.012356536, 0.006585338, -0.02406964, -0.0025770143, 0.01116159, -0.021062577, -0.019224199, -0.011752498, -0.030517101, 0.0070514986, -0.0050785234, 0.021574697, -0.016506022, 0.04687868, 0.009191898, 0.038527183, -0.002120702, -0.0018465866, 0.007931295, 0.00045590173, 0.04123223, -0.030359525, 0.03225043, -0.026971655, 0.0018597179, -0.0019073188, -0.0064376113, -0.02601307, -0.0062734704, -0.014601986, -0.00077310426, -0.021784797, -0.033405982, -0.0360585, -0.01821309, -0.002499868, 0.020826215, -0.04553929, 0.020484801, 0.00078090094, -0.031042352, 0.021745404, 0.012277749, 0.022152474, -0.013840372, 0.046826154, 0.04966251, 0.006020693, -0.025816102, 0.0074191745, 0.009500483, 0.0014034058, 0.019329248, 0.017779756, -0.0022142625, -0.013551484, -0.005419937, 0.004740393, -0.020051468, -0.042335253, 0.0027378723, -0.008167658, 0.0135252215, 0.007859073, 0.0149433995, -0.014772693, 0.0024785297, -0.003156432, 0.020773688, -0.0074454374, -0.005035847, 0.00091180345, 0.018239351, -0.0011186211, 0.0074388715, 0.0022536563, -0.016558547, -0.005268927, -0.008660081, 0.0034338303, 0.0017037839, 0.0017070668, 0.019079754, 0.0043267575, 0.023898935, -0.0046747364, -0.015705014, 0.0058270064, -0.0114636095, -0.013787847, 0.03471911, 0.0068348325, -0.026485797, 0.00027637248, 0.0030448162, 0.025422163, -0.03419386, -0.004087112, -0.016637335, -0.014470673, -0.0037095873, 0.027943369, 0.010642905, 0.015271681, 0.0047206962, -0.009507049, -0.0022864845, -0.021666616, -0.0055052903, 0.038133245, -0.046038277, 0.0009914119, 0.0009955154, 0.020865608, -0.0021732273, 0.027601955, 0.016939355, 0.009815633, 0.027995894, -0.036741327, 0.0026623677, -0.008758565, -0.008180789, -0.018370664, -0.0025491102, -0.012310578, -0.014956531, 0.0230454, -0.01208078, -0.02211308, 0.023465602, 0.0076161437, 0.016939355, 0.018501977, -0.010728258, 0.008069173, -0.004044435, 0.00409696, -0.01822622, 0.011200984, 0.0059583196, -0.014102997, -0.00019409676, 0.011279772, -0.019907024, -0.009382301, -0.012579769, -0.009664624, -0.019014098, -0.01203482, -0.018554503, -0.010143916, -0.00855503, -0.02467368, -0.001016033, -0.0017710817, -0.013630271, -0.025264587, 0.008804525, 0.0020763841, -0.010157047, -0.021207022, 0.017451474, 0.006000996, 0.0018679249, -0.008285839, -0.0125535065, -0.0046747364, -0.018593896, -0.027733268, 0.0052656443, 0.007281296, 0.047246356, 0.001352522, -0.048743322, -0.019696923, -0.023925196, -0.02928276, -0.014378754, -0.012015124, -0.013866634, -0.02723428, -0.014667642, -0.011004015, 0.004500747, 0.007300993, 0.024739336, 0.0059944303, -0.016506022, 0.019526217, -0.005239382, 0.020497931, -0.00012669637, -0.005672714, -0.012881788, 0.019841367, -0.007268165, -0.0041790307, 0.037030216, -0.01361714, 0.02057672, -0.0139322905, 0.012461587, 0.011601488, 0.017543394, 0.013906028, -0.0347979, -0.016821172, 0.022690857, 0.010104522, -0.025658526, 0.0050653922, -0.014457542, 0.020248437, -0.013709059, -0.0090802815, -0.01566562, 3.4315734e-05, -0.015901983, 0.0026738574, -0.007819679, 0.0043628686, 0.016781779, 0.031751443, 0.0029578214, -0.0035257493, 0.003972213, 0.017188849, -0.00010658908, 0.026078727, 0.008784828, -0.008817656, -0.02282217, -0.011437347, -0.017832281, -0.022914087, -0.020852476, -0.0041691824, 0.008666647, 0.017753495, -0.00012064366, -0.020222176, -0.008207052, 0.010124219, 0.009356039, -0.010590379, -0.006060087, -0.00890301, -0.014142391, -0.012074214, 0.0045598377, -0.014733299, -0.02765448, 0.011141893, 0.0017612333, -0.018541371, 0.04167869, 0.21020557, -0.0056661484, 0.03571709, -0.0018580764, 0.0009987982, 0.0135252215, -0.004976756, 0.002215904, 0.010839874, -0.008174224, 0.013853503, 0.010262097, -0.004999736, 0.0006577952, 0.005193422, -0.027391855, -0.034535274, 0.0034010021, -0.03550699, -0.015849458, 0.0027395138, -0.0011284696, -0.0032105986, -0.013380777, 0.00977624, -0.019605005, -0.030727202, 0.0006118357, 0.008154526, -0.004241404, -0.01566562, 0.00076571794, 0.0039492333, -0.014864611, -0.0021896414, 0.0048224637, -0.013420171, -0.016151477, 0.022060554, -0.009612099, 0.013315121, 0.0123828, 0.011187852, -0.01864642, 0.006605035, -0.008660081, 0.011036843, 0.0070908926, -0.010518157, 0.019014098, -0.031410027, -0.00537726, 0.020103993, 0.021994898, -0.027680743, -0.01780602, 0.008686343, 0.023426209, -0.001216285, 0.026393877, 0.01586259, 0.015363601, -0.034955475, 0.017976725, -0.02212621, 0.018318139, -0.028810034, 0.007149983, 0.017635312, -0.022296919, -0.003552012, -0.02529085, 0.006628015, 0.021469647, -0.02435853, -0.018607028, 0.048848372, 0.0014813729, 0.004408828, 0.03335346, -0.004192162, -0.011516135, -0.042230204, -0.017412081, -0.00411994, -0.017359555, 0.015219157, 0.012028255, -0.039761525, -0.01566562, 0.028232258, -0.007412609, -0.008870182, -0.0052459473, -0.0037588296, -0.00054453785, -0.017674707, 0.015245419, -0.00772776, 0.0047075646, -0.01080048, 0.07810487, -0.017976725, -0.0019007531, -0.0046025147, -0.0002033297, -0.0011957674, 0.010373713, -0.020865608, -0.015258551, 0.0017218394, -0.03658375, 0.008883312, -0.012330274, -0.0023143885, 0.027995894, 0.017018143, -0.03295952, -0.00921816, 0.0074257404, -0.020287832, -0.0061027636, 0.020051468, 0.025947414, -0.016112084, 0.004116657, -0.004241404, -0.009277252, -0.01647976, -0.031882755, 0.034824163, 0.00885705, 0.010872702, 0.02652519, -0.012908051, 0.007596447, 0.014208048, -0.006381803, -0.016322184, 0.02354439, -0.019815106, -0.026341353, -0.034745373, 0.0057744817, 0.003696456, 0.009040888, 0.0123828, 0.011292903, -0.0079050325, 0.004871706, -0.039577685, -0.01577067, 0.0039098393, 0.004280798, 0.018961571, -0.02865246, -0.009690886, -0.025842365, 0.0074388715, 0.012185831, -0.021141365, -0.014155523, -0.013229767, -0.012691384, -0.0018531523, -0.01315098, -0.1661895, -0.010071694, -0.0018121169, -0.023176713, 0.009992906, 0.02313732, 0.0007296069, -0.01934238, 0.015691882, -0.0013418529, 0.019880762, -0.0021075709, -0.014431279, -0.038474657, 0.017700968, 0.012494415, -0.022690857, 0.027496906, 0.018961571, -0.0069661452, 0.0071105896, -0.009533311, 0.014470673, 0.0038343344, -0.00037034322, 0.022152474, -0.011765629, -0.014299966, -0.002808453, 0.010176744, -0.02959791, 0.017136324, 0.029729223, 0.017569656, 0.013269161, -0.014693906, -0.008128264, 0.016506022, -0.015429257, 0.043674644, 0.035480727, 0.012455022, -0.0042381217, -0.011365125, -0.014483805, 0.016519153, 0.017740363, 0.01504845, 0.027523167, 0.004415394, 0.013682797, 1.596528e-05, 0.019289855, 0.019355511, 0.016243396, -0.005781047, -0.012107043, 0.023741359, -0.00657549, 0.014352492, 0.0033583255, -0.017871676, 0.016151477, -0.025658526, 0.015179763, -0.021902978, -0.003368174, 0.013308555, -0.02804842, 0.0048946855, -0.015376732, -0.027917108, 0.0026705747, -0.008285839, -0.000280476, 0.004405545, -0.021732273, 0.026919128, 0.012461587, 0.0107348235, -0.005639886, 0.014510067, -0.0036373653, -0.0026558018, -0.01638784, -0.010255531, -0.008029779, -0.0029742355, -0.011128762, -0.015823195, 0.018895915, -0.022204999, 0.01084644, 0.0029299175, 0.041100916, 0.01229088, -0.019880762, -0.015705014, 0.0016808042, 0.016729254, 0.026131252, -0.0062143793, 0.008935838, 0.023596914, 0.015810065, 0.0056464514, 0.007117155, -0.0045532724, 0.043359496, -0.021430254, 0.00033115453, -0.006769176, 0.023518126, 0.029098922, -0.0065262476, 0.03513931, -0.00036747073, -0.006992408, 0.004376, -0.020261569, 0.06324026, 0.027418118, 0.0027789078, 0.025776708, -0.031803966, -0.011594922, -0.0911311, -0.02928276, 0.0023833278, 0.015284813, -0.007333821, 0.0014789107, -0.006345692, 0.0076818005, 0.003604537, 0.0064573083, 0.005219685, -0.022388836, -0.032696895, -0.016663598, 0.010583813, 0.010189875, -0.010912096, -0.013866634, -0.007327256, 0.025146406, 0.0049209483, 0.01171967, -0.01709693, -0.0170838, 0.022401968, 0.014418148, -0.034535274, -0.0049734735, 0.02160096, 0.00591236, 0.0011210833, -0.022756513, 0.023150451, 0.004799484, -0.008837353, -0.0051113516, 0.014444411, -0.01626966, 0.029335285, -0.00804291, 0.007117155, -0.002288126, 0.009336342, -0.0035585775, 0.0066706915, 0.0008005979, 0.006545944, 0.0013475979, 0.016519153, -0.0057186736, -0.036163554, -0.019775711, -0.02682721, -0.0041133743, 0.02273025, -0.04036556, -0.007012105, 0.006421197, -0.01336108, -0.027050443, -0.038553447, 0.001876132, -0.009073717, 0.03143629, 0.022467624, -0.01667673, -0.031173665, -0.038448397, 0.027969632, -0.014405017, -0.0150615815, 0.032302953, -0.0025540343, 0.016873699, -0.041442327, -0.008706041, -0.03288073, -0.027575694, 0.027995894, -0.032801945, -0.004083829, -0.03603224, 0.024240348, -0.021075709, 0.018344402, 0.0026672918, 0.0131641105, -0.004812615, 0.0030333262, -0.0018826977, -0.008233314, -0.0041133743, 0.01182472, -0.01028836, 0.0037456984, 0.01233684, 0.00160612, 0.0016414103, -3.444397e-05, 0.022296919, -0.03603224, -0.006992408, -0.04393727, 0.023623178, -0.00911311, -0.001252396, 0.011516135, -0.012638859, 0.03274942, 0.007360084, 0.021771666, 0.0155474385, -0.00499317, 0.01914541, -0.017228242, -0.0064179143, 0.009992906, 0.012218659, 0.033721134, 0.020025207, 0.039787788, 0.023465602, -0.005177008, 0.019092884, 0.012967141, 0.015796933, -0.016190872, -0.0022224695, -0.016729254, 0.004467919, -0.0015946301, 0.015402994, 0.018541371, -0.03298578, -0.0074913967, 0.004799484, -0.0064179143, -0.012310578, -0.025776708, 0.020090863, 0.024003984, 0.006421197, -0.019289855, -0.02200803, 0.0023143885, -0.00732069, 0.021312071, 0.020353489, -0.023833279, -0.005410088, -0.002964387, -0.012126739, 0.015534307, 0.018002989, 0.0024374942, -0.010912096, -0.014720168, -0.011982296, 0.013367645, -0.020642376, 0.029650437, -0.012277749, -0.00147809, 5.1191484e-05, 0.0011284696, -0.0079050325, -0.010629773, 0.0033944366, -0.015705014, -0.00015624175, 0.0012679895, -0.0401292, -0.027575694, 0.009546442, -0.021259546, 0.021233283, 0.018922178, -0.00012608083, -0.027733268, 0.007543922, -0.0034206992, -0.005177008, 0.023268633, -0.00028355364, -0.022283787, 0.0116277505, 0.0150221875, 0.007353518, 0.004228273, 0.0054265023, 0.015626226, 0.0052426644, -0.0035257493, 0.019644398, 0.004241404, 0.00062004273, 0.024148429, 0.0013369287, -0.019605005, -0.014352492, 0.004067415, 0.024594892, -0.026052464, -0.012546941, -0.017044405, -0.015705014, -0.011457044, 0.02477873, -0.01382724, -0.031278715, -0.013032798, -0.018830258, -0.003988627, -0.01956561, -0.010413107, 0.00027062753, 0.006559076, 0.007012105, 0.010971187, 0.0149433995, -0.008463112, 0.04209889, 0.00881109, 0.023912065, 0.0015872438, -0.0025835799, 0.035585776, 0.029125184, 0.012507547, -0.014155523, 0.047587767, -0.029151447, 0.011496438, 0.008174224, -0.02804842, 0.00028622092, -0.0038999908, 0.0154686505, -0.003643931, 0.021666616, -0.010419672, 0.07463821, -0.011798457, -0.025461556, -0.014194916, -0.009631796, 0.004812615, -0.011765629, 0.0149433995, -0.035664562, -0.015888851, -0.0006212738, -0.028888822, 0.027785795, -0.039971624, 0.010701995, -0.0085419, -0.017884808, 0.011516135, 0.006591904, -0.010012603, 0.018987834, -0.0023734793, 0.0048815543, -0.0011416009, -0.042834245, -0.005547967, 0.036216076, 0.0023587067, 0.020117125, -0.038763545, 0.008174224, 0.030779727, -0.02670903, -0.001876132, 0.025789838, 0.0038704455, -0.007281296, -0.0078656385, 0.044882722, 0.0132428985, -0.0021551717, -0.007209074, -0.018108038, -0.019775711, 0.0065262476, 0.0037653954, -0.0059254915, -0.021338334, -0.015586833], 'triplet_index': 0, 'head_id': 'PHYHIP_(0)', 'tail_id': 'KIF15_(8889)', 'edge_type': 'gene/protein|ppi|gene/protein'}"]

In [9]:

# Check the ground truth for the search
results[0]['triplet_index'], results[0]['head_id'], results[0]['tail_id'], results[0]['edge_type'], results[0]['feat']

# Check the ground truth for the search results[0]['triplet_index'], results[0]['head_id'], results[0]['tail_id'], results[0]['edge_type'], results[0]['feat']

Out[9]:

(0,
 'PHYHIP_(0)',
 'KIF15_(8889)',
 'gene/protein|ppi|gene/protein',
 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with KIF15 (gene/protein).')

In [10]:

import numpy as np

import numpy as np

In [11]:

%%time

# Assume node_coll_name is defined and collection exists
collection = Collection('t2kg_primekg_edges')

# Load the collection into memory before query
collection.load()

# Vector similarity search in Milvus
vector_to_search = normalize_vector(np.array(results[0]['feat_emb'])).tolist() # merged_edges_df["edge_emb"].iloc[0]
search_params = {"metric_type": "IP"}
results = collection.search(
    data=[vector_to_search],
    anns_field="feat_emb",
    param=search_params,
    limit=10,
    output_fields=["head_id", "tail_id", "edge_type", "feat"]
)
results

%%time # Assume node_coll_name is defined and collection exists collection = Collection('t2kg_primekg_edges') # Load the collection into memory before query collection.load() # Vector similarity search in Milvus vector_to_search = normalize_vector(np.array(results[0]['feat_emb'])).tolist() # merged_edges_df["edge_emb"].iloc[0] search_params = {"metric_type": "IP"} results = collection.search( data=[vector_to_search], anns_field="feat_emb", param=search_params, limit=10, output_fields=["head_id", "tail_id", "edge_type", "feat"] ) results

CPU times: user 509 ms, sys: 154 ms, total: 664 ms
Wall time: 1.05 s

Out[11]:

data: [[{'triplet_index': 0, 'distance': 1.000000238418579, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with KIF15 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'KIF15_(8889)'}}, {'triplet_index': 3069556, 'distance': 0.9815623760223389, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'KIF15 (gene/protein) has a direct relationship of protein_protein:ppi with PHYHIP (gene/protein).', 'head_id': 'KIF15_(8889)', 'tail_id': 'PHYHIP_(0)'}}, {'triplet_index': 93582, 'distance': 0.9684444665908813, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with PRKD2 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'PRKD2_(9221)'}}, {'triplet_index': 305788, 'distance': 0.9657970666885376, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with PPIE (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'PPIE_(5629)'}}, {'triplet_index': 39303, 'distance': 0.9635176658630371, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with PAQR5 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'PAQR5_(8043)'}}, {'triplet_index': 300682, 'distance': 0.9599261283874512, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with PRMT5 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'PRMT5_(7026)'}}, {'triplet_index': 3154085, 'distance': 0.9597105979919434, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with SUPT5H (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'SUPT5H_(1736)'}}, {'triplet_index': 3201926, 'distance': 0.9593513607978821, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with S100A13 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'S100A13_(5140)'}}, {'triplet_index': 3334423, 'distance': 0.9588445425033569, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with HNRNPA1 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'HNRNPA1_(352)'}}, {'triplet_index': 110476, 'distance': 0.9586011171340942, 'entity': {'edge_type': 'gene/protein|ppi|gene/protein', 'feat': 'PHYHIP (gene/protein) has a direct relationship of protein_protein:ppi with ANKRD12 (gene/protein).', 'head_id': 'PHYHIP_(0)', 'tail_id': 'ANKRD12_(6038)'}}]]

In [12]:

# Get node indices from the results
[n['triplet_index'] for n in results[0]]

# Get node indices from the results [n['triplet_index'] for n in results[0]]

Out[12]:

[0, 3069556, 93582, 305788, 39303, 300682, 3154085, 3201926, 3334423, 110476]

In [13]:

# Get the cosine similarity scores
[n['distance'] for n in results[0]]

# Get the cosine similarity scores [n['distance'] for n in results[0]]

Out[13]:

[1.000000238418579,
 0.9815623760223389,
 0.9684444665908813,
 0.9657970666885376,
 0.9635176658630371,
 0.9599261283874512,
 0.9597105979919434,
 0.9593513607978821,
 0.9588445425033569,
 0.9586011171340942]

In [ ]: