23 Feb 2021   |   Network Updates   |   Update from University of Tartu
These updates are republished press releases and communications from members of the Science|Business Network

University of Tartu: international consortium uses first US-European genomic and health data set for AI-supported personalised medicine


The University of Tartu is participating in an international project INTERVENE that helps to create next-generation methods for the prevention, diagnosis and personalised treatment of diseases. To this end, the first US-European genomic and health data set is used, which, connected to an artificial intelligence-supported data analysis platform, provides a great opportunity for genetic data to reach a wide range of clinical use. 

One of the cornerstones of personalised medicine is the system of risk scores that combine genetic, environmental and lifestyle risk factors. They allow a more accurate prediction of the development of various common and rare diseases. In current medical practice, clinical risk assessment is used to predict the development of diseases. However, the prognosis would be even more accurate if the person’s genetic data were also taken into account alongside clinical factors. 

The international consortium led by the University of Helsinki brings together 17 partners from all over Europe as well as the US to create a platform for the wider use of genetic data that considers the legal and ethical framework involved. 

In the project, the University of Tartu deals with harmonising the data of biobanks participating in the project and takes part in a clinical pilot project. According to the Estonian coordinator of the project, Professor in Bioinformatics of the University of Tartu Reedik Mägi, the application of artificial intelligence on big data requires transferring the health and gene data of all the biobanks involved to a comparable machine-readable format. “Researchers of the University of Tartu Institute of Computer Science have already started with this work and we hope that we can also share our experience with other project partners,” said Mägi. 

The University of Tartu also participates in a clinical pilot project which assesses the impact of genomic data-based advice on people at high risk of cardiovascular disease. “Together with researchers of the University of Tartu Institute of Family Medicine, we have previously carried out several pilot projects dealing with advice based on genetic risk scores and its impact. One of the greatest strengths of our biobank is that people can be easily called back both for counselling and collecting additional information,” Mägi explained.

In addition to known clinical factors and the genetic map, the project uses as many other types of health data of the person as possible to create the risk scores. Artificial intelligence-based methods are used to develop the best possible models using a large amount of genomic and health data. The models are then tested to predict the risk of cardiovascular disease, type 2 diabetes and breast cancer based on the biobanks in Estonia, Finland and Italy.

A next-generation healthcare service

In cooperation with experts from different fields, a pan-European platform will be built to allow researchers and doctors to easily calculate the risk scores of different diseases. ”This would make it possible to extend the use of the method in clinical practice. Despite increasing evidence of the effectiveness of genetic risk scores, they have not been introduced in the clinical practice for many diseases,” explained the Estonian coordinator of clinical studies, Research Fellow of Public Health of the University of Tartu Mikk Jürisson.

In the course of the project, a randomised clinical study will be conducted among young adults at high risk of cardiovascular disease. The study assesses whether early knowledge of a high genetic risk leads people to make healthy choices and reduces disease risk. If this hypothesis is confirmed, the European Society of Cardiology will be invited to include genetic risk scores in clinical guidelines. “We hope that as a result of the project we can develop reliable and effective methods for preventing different diseases,” said Jürisson.

Reedik Mägi added that the novel tools to be developed are expected to automatically calculate the risk scores for people of different ancestry in many countries and to communicate this information to doctors and patients in an understandable manner.

The international consortium for integrative genomics prediction (INTERVENE) has received 10.4 million euros from the European Commission’s Horizon 2020 programme (grant agreement no. 101016775). The leading partner of the project is the University of Helsinki. The project will run until the end of 2025. 

Never miss an update from Science|Business:   Newsletter sign-up