chapter four

4 From simple networks to multisource integration

 

This chapter covers

  • Building and integrating complex knowledge graphs
  • Exploring examples of knowledge graphs
  • Understanding analysis and query techniques
  • Analyzing KG results with LLMs

This chapter extends our understanding of how to construct increasingly large and complex knowledge graphs (KGs) and use them to develop intelligent advisor systems (IASs). Whereas in chapter 3, we had a single knowledge base in the form of an ontology, from this point on, we will create KGs from multiple structured data sources that are available in graph-friendly formats. This approach will let us focus on graph modeling decisions, integration strategies, and analysis methods.

Note  Appendix C on the book’s website offers comprehensive guidance on ingesting and transforming raw data from multiple complex sources.

The examples in the following sections cover transforming structured and semistructured schemas and data formats into a homogeneous graph, reconciling and matching names and identifiers, post-processing techniques to merge entities and relationships, and analyzing the resulting KG to find relevant information. We use biomedical data sources, but the techniques and patterns are directly transferable to other domains.

4.1 Biomedical knowledge graphs and applications

4.2 Multi-omic applications of KGs

4.2.1 Creating a KG from the PPI and protein-disease networks

4.2.2 High-level analysis of the resulting KGs

4.2.3 Domain-specific analysis of the PPI and disease KG

4.3 Pharmaceutical applications of KGs

4.3.1 Deep analysis of the Hetionet knowledge graph

4.3.2 LLM-assisted interpretation of pathway analysis results

4.4 Clinical applications of KGs

4.4.1 LLM-guided clinical decision support analysis

Summary