DbKB a knowledge graph dataset for diabetes: A system biology approach

Area of Science:

• Genomics and Bioinformatics
• Systems Biology
• Metabolic Diseases

Background:

• Diabetes is a widespread disease affecting millions globally.
• Identifying key genes is crucial for understanding diabetes pathogenesis.
• Type 2 diabetes (T2D) involves complex genetic and molecular interactions.

Purpose of the Study:

• To develop a multi-layer Knowledge Graph (DbKB) for visualizing and querying diabetes-related biological networks.
• To identify regulatory genes, functional genes, and protein-protein interactions in T2D.
• To facilitate the study of interactions among T2D candidate genes.

Main Methods:

• Construction of a multi-layer Knowledge Graph (DbKB) based on the Multi-Scale Network Model for Diabetes (MSNMD).
• Integration of gene regulatory networks and protein-protein interaction networks.
• Identification of regulatory genes from the TRRUST database and functional genes involved in T2D.

Main Results:

• The DbKB integrates genetic, molecular, and regulatory information for T2D.
• Identified 875 regulatory transcription factors in the first layer and 550 regulatory genes in the second layer.
• Facilitated the identification of interaction-driven sub-networks involving regulatory and functional genes.

Conclusions:

• The developed Knowledge Graph provides a novel approach to analyze complex T2D networks.
• DbKB enables deeper insights into gene regulation and protein interactions in diabetes.
• This tool aids in scrutinizing T2D candidate genes and understanding disease mechanisms.