Headache disorders, including migraine, are common and debilitating conditions that affect millions of people worldwide. Headache disorders are associated with severe pain, nausea, sensitivity to light and sound, and commonly have a major impact on daily activities and thus quality of life. Although there have been recent breakthroughs in the management of migraine, critical medical need remains to identify effective and tolerable treatment options for migraine and other headache disorders.
Lundbeck believes that the key to developing new and better therapies for headache disorders lies in understanding the genes and biological pathways that are involved in these conditions. However, identifying the right drug targets by evaluating biological information is not easy. The molecular pathophysiology of headache disorders is complex and poorly understood; there are thousands of possible genetic links to these diseases and many biological pathways to consider.
Traditionally, researchers spend a lot of time and money searching through scientific literature and subsequently conducting experiments in the lab to identify and validate potential drug targets. This process is time-consuming and costly and Lundbeck wanted to find a more efficient way to aid the identification of new drug targets for headache disorders as well as supporting existing programs with the best possible and structured information.
Lundbeck decided to use a combination of knowledge graphs, a way to identify and analyze the link between different data sets, and AI to augment its research process and generate hypotheses in silico, meaning using computational models instead of physical experiments. To achieve this, Lundbeck partnered with Accenture, a global leader in AI and data science.
Together, they built a domain-specific knowledge graph for headaches and migraines, using Accenture’s large-scale clinical dataset. Their dataset consists of 54 million electronic medical records that include information on clinical characteristics and public data sources including biological pathways, genetics, and diseases. A knowledge graph is a network of entities and relationships that captures the relevant information in a structured and semantic way. For example, a knowledge graph can link genes, diseases, symptoms, drugs, pathways, and other biological and clinical concepts.
Accenture then applied machine learning models to predict the probability of gene-disease links in the knowledge graph, based on the available evidence. These models can learn from the existing data and infer new connections that are not explicitly stated. For example, if a gene is known to be involved in a pathway that is related to a disease, the model can suggest that the gene is also associated with the disease.
Using AI, Lundbeck was able to find potential drug targets associated with headache disorders, such as migraine, 80% faster than traditional methods. Furthermore, using the knowledge graph machine learning models, the team demonstrated that the approach could identify 77% of known gene-headache associations and suggested novel biological targets for drug discovery.
As a crucial aspect of the project, the Accenture team also provided explanations for their predictions, showing Lundbeck the key facts in the knowledge graph that influenced the decisions. This helped Lundbeck to understand genetic and biological links and build pathophysiological hypotheses through the use of AI, thereby prioritizing the most promising ones for further validation.
By using AI, Lundbeck and its partner Accenture developed a new approach to interrogating biological datasets that are now being widely used in Lundbeck’s Research departments to aid in the generation of new ideas and substantiating links between potential new drug targets and diseases. In addition to providing a more efficient framework for daily research use, Lundbeck has gained a deeper understanding of the use of AI and knowledge graphs in preclinical drug discovery, and how it can complement existing expertise and methods.
In summary, AI is helping Lundbeck to advance our neuroscience research and development process, reduce costs, and potentially discover new treatments that can improve the lives of millions of people living with headaches and migraines.