Can Knowledge Graphs Revolutionise Pharma R&D?

In life sciences, the adoption of new standards such as Study Data Tabulation Model (SDTM) and Analysis Data Model (ADAM) is proving critical for efficient and effective data management and sharing. SDTM provides a new way of organising human clinical and nonclinical study data tabulations, which is required for data submission to regulatory bodies like the United States Food and Drug Administration (FDA) and Pharmaceuticals and Medical Devices Agency (PMDA) Japan, while ADAM defines dataset and metadata standards for clinical trial statistical analyses, ensuring efficient generation, replication, and review of data. The Clinical Data Interchange Standards Consortium (CDISC) 360 is another important initiative that aims to implement standards as linked metadata to support metadata-driven automation across the entire clinical research data lifecycle, making it easier for researchers to analyse and share their findings. The digital transformation of pharma industry regulatory processes, hence, has started to make data a key tool. However, with the increasing volume and complexity of data generated in drug discovery and clinical research, life sciences R&D practitioners need better ways of organising, structuring and exploiting their data, at scale.

To address the increasing volume and complexity of data generated in drug discovery and clinical research, the sector is increasingly adopting an innovative data structure approach, the knowledge graph. Graph databases can tackle complex problems in drug discovery, multiomics, and clinical research by allowing researchers to store and analyse complex interconnected data such as relationships between genes, proteins, cells, and tissues, as well as help the sector get better at meeting standards like SDTM and AdaM.

The main advantage knowledge graphs offer is their basic design. Unlike traditional SQL databases that use fixed tables with rows and columns to store data, knowledge graphs represent data as interconnected ‘nodes’ (or entities) linked by ‘edges’ (or relationships).

This network (a graph is a mathematical name for a network) of interconnections holds the key to unlocking breakthrough insights. The power of knowledge graphs is evident in their ability to represent complex data relationships. In the Panama Papers work, for example, a knowledge graph helped uncover an intricate network of opaque offshore accounts, shell companies, and individuals allowing investigators to connect the dots and uncover hidden relationships. These insights would have been difficult to detect using traditional data analysis methods.

Owing to their ability to represent intricate data, knowledge graphs have many applications beyond financial investigations. One such area is biological science, where knowledge graphs can capture the intricate interconnections and correlations among diseases, genes, environment, diet, behaviour, and other factors.