BioDWH2

BioDWH2 provides graph-based data warehousing and mapping to integrate and cross-reference heterogeneous biomedical datasets for analysis of proteomics, pharmacogenomics, and foodomics.

Key Features:

• Graph-based data warehouse: Stores integrated data as a graph structure to represent entities and relationships across sources.
• Mapping between datasets: Performs entity mapping to link equivalent or related items across heterogeneous databases.
• Workspace-centered source selection: Uses a workspace model to select and configure which data sources are included in a project.
• Heterogeneous data format integration: Supports ingestion and integration of diverse data formats from multiple sources.
• Neo4j and GraphQL access: Exposes the integrated graph via Neo4j or a GraphQL server for querying and programmatic access.
• Cross-database entity cross-referencing: Enables cross-referencing of entities from different databases to identify links and relations.
• Support for large-scale omics: Targets consolidation and analysis of large-scale omics datasets, including proteomics, pharmacogenomics, and foodomics.

Scientific Applications:

• Proteomics integration: Consolidates proteomics datasets to enable cross-referenced analysis of proteins and interactions.
• Pharmacogenomics mapping: Integrates pharmacogenomics data to link genetic variants, drugs, and phenotypic effects.
• Foodomics data synthesis: Aggregates foodomics datasets to relate dietary compounds to biological entities and pathways.
• Network-based relationship discovery: Identifies relationships and patterns across diverse biological data sources using graph structures.

Methodology:

Constructs a graph-based data warehouse, applies a workspace-centered selection of data sources, integrates heterogeneous data formats, and exposes the integrated graph via Neo4j or a GraphQL server for querying and cross-referencing.