A New Tool For Understanding Inherited Heart Disease
A new online tool for interpreting genetic data involved in inherited heart disease has been developed by the Cardiovascular Genetics and Genomics (CGG) team at Imperial College London and the Royal Brompton Hospital. The tool, known as CardioClassifier, is designed to produce fast, high-quality results for clinically diagnosing heart disease.
“This web-based tool uses automatic processes to help us to focus in on the DNA changes that really matter to patients and to interpret these changes accurately using huge amounts of data,” said Stuart Cook, Ph.D., Professor of Clinical and Molecular Cardiology and Head of the Myocardial and Vascular Biology Division for NHLI. “It also brings variant interpretation to the clinician and helps disseminate genetic testing for the benefit of all, wherever you are in the world.”
More than 600,000 people in the UK alone are affected by inherited heart conditions, and healthcare organisations such as the NHS are increasingly using genetic testing to identify and diagnose new cases. Understanding the genetic roots of these diseases is an important step in identifying novel treatment pathways, as well as helping clinicians to extend their help to family members of patients who might also have inherited the condition.
However, it’s not as simple as it might initially appear. As with other next generation sequencing projects, clinicians are currently being faced with the multitude of challenges presented by big data which they cannot easily overcome. Storage and transportation of medical data can be a problem on a large scale, but one of the more significant barriers for widespread adoption is data interpretation. Any patient’s genome will contain a vast array of variants, only a small number of which will be linked to their cardiac health, and identifying then interpreting this subset of data can be a very difficult and time consuming task.
CardioClassifier has been designed to solve these interpretation challenges. By focusing solely on cardiac disease, the tool can offer better performance than other, less specific interpretation platforms and interpretation accuracy has been improved significantly through expert reviews.
“When you look at how many large genome projects work, they’re about genome-wide, one-size-fits-all solutions because they’re big and scalable,” said James Ware, Ph.D., group head and clinical senior lecturer, when speaking with FLG. “But when it comes to clinical application, you really need to optimise things for particular diseases or particular genes. You need more specific knowledge. No one team can have that across the genome.”
To use the system, the user has to upload their list of genetic variants and select which cardiac disease they wish to focus on. The program will then evaluate the variants, decide which ones are relevant for the disease in question, and automatically annotate them with evidence to support their importance (in accordance with ACMG guidelines). The results of this interpretation can be viewed via an interactive web page, and clinicians can add any further information that they may have about their patient to generate an integrated, definitive result.
CardioClassifier has been linked with a number of variant databases, including ClinVar, to increase the depth of variant data available, but it has also been designed to grow over time. When users add their own information to the program, the system can use that new data to learn about more diverse or rare variants.
“It will learn. Of course, we’ll populate it with data as data grows, but it can use information uploaded by our users,” said Dr. Ware. “If you use the free licence, then you’re signing up to the anonymous data being available to others. As people are using the tool, it will learn what people are saying about variants and what evidence they’re using.”
The program should not only save time for clinicians, but it integrates the results in a standard, reproducible way to make them as user-friendly as possible.
When asked about the transferability of the program to other diseases or conditions, Dr. Ware told us that, while CardioClassifier was a specific tool, the highly-focused approach was certainly something which could be useful in other clinical areas. “We have ambitions to be more re-deployable in some of our on-going work and I think that’s certainly the direction of travel,” he said.
Beyond CardioClassifier, the CGG team have also been developing a range of tools to better enable genomic testing within the clinical setting, such as the TruSight Cardio Sequencing Kit to sequence important heart genes. The Wellcome Trust, the Department of Health, the Medical Research Council, and the British Heart Foundation are all involved in supporting their work.