Periodic Reporting for period 2 - INTERVENE (International consortium for integrative genomics prediction)
Reporting period: 2022-01-01 to 2023-06-30
Data science and the adoption of artificial intelligence (AI) -based methods is expected to transform our ability to understand disease development and trajectories. It is also expected to generate new treatment options that are tailored toward the specific genetic and other characteristics of the patient. Nonetheless, there remains several and significant challenges to the advance of data science and translation into clinical practice.
Though many diseases have an underlying genetic background, it is uncommon that a particular disease can be straightforwardly traced to one or a few individual genes. Instead, the risk of developing a disease, and the way the disease progresses and is susceptible to treatment, is typically affected by many parts of the genome as well as non-genetic factors acting together, and in which the contribution of individual genes may be only very small.
Methods for genomic sequencing and molecular profiling together with digitalization of health care data collections have opened opportunities for more personalized disease care. One opportunity lies in the development of genetic risk scores, which summarize the estimated effect of an individual’s genetic makeup on the risk of developing a particular disease. In essence, the scores provide an estimate for the cumulated contributions of an individual’s genome for a specific disease or other outcome.
INTERVENE is an international consortium that seeks to advance AI-facilitated analyses of complex medical data to develop genetic risk scores for improved understanding of diseases and treatment options tailored to individuals. Utilizing unique data resources of genomic and other health information in international large-scale, multi-ethnic biobank projects, INTERVENE seeks to test and produce risk scores with improved predictive value for complex and rare diseases, applicability for disease screening, and comprehensibility for clinicians and citizens. The risk scores are developed using sophisticated computational algorithms that condense information from genetic variants and other health data into a number that reflects a person’s inherited susceptibility toward a disease.
INTERVENE seeks to meet these aims by:
1. Harmonising data on more than 1.7 million genotyped / sequenced genomes with longitudinal clinical data and facilitating access to summary statistics for comparative meta-analysis.
2. Developing AI-based integrative risk scores for the next generation of predictive and personalised medicine with continuous improvement and disease coverage of these scores
3. Testing the utility of novel risk scores in clinical settings and developing a strategy for their implementation in Europe and beyond.
Though many diseases have an underlying genetic background, it is uncommon that a particular disease can be straightforwardly traced to one or a few individual genes. Instead, the risk of developing a disease, and the way the disease progresses and is susceptible to treatment, is typically affected by many parts of the genome as well as non-genetic factors acting together, and in which the contribution of individual genes may be only very small.
Methods for genomic sequencing and molecular profiling together with digitalization of health care data collections have opened opportunities for more personalized disease care. One opportunity lies in the development of genetic risk scores, which summarize the estimated effect of an individual’s genetic makeup on the risk of developing a particular disease. In essence, the scores provide an estimate for the cumulated contributions of an individual’s genome for a specific disease or other outcome.
INTERVENE is an international consortium that seeks to advance AI-facilitated analyses of complex medical data to develop genetic risk scores for improved understanding of diseases and treatment options tailored to individuals. Utilizing unique data resources of genomic and other health information in international large-scale, multi-ethnic biobank projects, INTERVENE seeks to test and produce risk scores with improved predictive value for complex and rare diseases, applicability for disease screening, and comprehensibility for clinicians and citizens. The risk scores are developed using sophisticated computational algorithms that condense information from genetic variants and other health data into a number that reflects a person’s inherited susceptibility toward a disease.
INTERVENE seeks to meet these aims by:
1. Harmonising data on more than 1.7 million genotyped / sequenced genomes with longitudinal clinical data and facilitating access to summary statistics for comparative meta-analysis.
2. Developing AI-based integrative risk scores for the next generation of predictive and personalised medicine with continuous improvement and disease coverage of these scores
3. Testing the utility of novel risk scores in clinical settings and developing a strategy for their implementation in Europe and beyond.
The INTERVENE project has now completed its first half of its scheduled activities. In the first 30 months, the consortium has clearly met its targets. We have accomplished most of the deliverables staked out in our strategic roadmap for the first half of the project period and are embarking on the second half well positioned to meet our stated objectives to significantly advancing understanding and treatment of disease by leveraging AI-enabled analyses large genomic and healthcare data to produce the next generation of disease risk scores.
The scientific output from INTERVENE has also been extensive with more than 40 papers published thus far, including Identification of genetic risk factors (including PRS) impacting healthy life years (Jukarainen et al 2022 Nature Medicine), the identification of interpretable gene-biomarker associations in 190,000 genomes (Remo et al 2022 Nature Communications) and the systematic comparison of family history and polygenic risk across 24 common diseases (Mars et al 2022 AJHG). Through ongoing clinical studies in Estonia, Finland, and Italy, we are launching major lines of enquiry to assess to what extent awareness of one’s genetic risk can influence preventive health behavior in cardiovascular disease and breast cancer. We are also continuing building a solid framework for the ethical and regulatory acceptance of using AI-derived (genetic) prediction scores for supporting clinical research and decision making in collaboration with patient organizations, and ethics and regulation experts. In parallel, INTERVENE has worked towards creating a cloud-based platform that can be used by the scientific community to expedite the calculation and interpretation of next-generation polygenic scores. The development of this platform at its integration with the PGS catalog (Polygenic Score Catalogue) has been developed and it is currently in advanced testing.
Following the project launch during the global challenge posed by the coronavirus pandemic, we have taken active measures to strengthen connections and collaborations across INTERVENE through in-person meetings, workshops, and to strengthen dissemination of the project, for example at international conferences. INTERVENE project postdocs and PIs have also been presenting the project and its results at the leading genetics conferences including American Society of Human Genetics (ASHG) and European Society of Human Genetics (ESHG) conferences. At the ASHG, INTERVENE was featured in a high-profile promotional video of the project in its Thought Leadership Series. INTERVENE partners have also individually and in partnerships with similar initiatives actively participated in promoting the translation of PRS into health care, for example through our ongoing clinical studies, in parallel with developing consensus guideline for trustworthy and deployable artificial intelligence in healthcare, and describing the gaps in knowledge for PRS translation.
Overall, we are moving along our projected track for reaching our objectives and to help realize the next generation of polygenic risk scores for the benefit of science and society.
The scientific output from INTERVENE has also been extensive with more than 40 papers published thus far, including Identification of genetic risk factors (including PRS) impacting healthy life years (Jukarainen et al 2022 Nature Medicine), the identification of interpretable gene-biomarker associations in 190,000 genomes (Remo et al 2022 Nature Communications) and the systematic comparison of family history and polygenic risk across 24 common diseases (Mars et al 2022 AJHG). Through ongoing clinical studies in Estonia, Finland, and Italy, we are launching major lines of enquiry to assess to what extent awareness of one’s genetic risk can influence preventive health behavior in cardiovascular disease and breast cancer. We are also continuing building a solid framework for the ethical and regulatory acceptance of using AI-derived (genetic) prediction scores for supporting clinical research and decision making in collaboration with patient organizations, and ethics and regulation experts. In parallel, INTERVENE has worked towards creating a cloud-based platform that can be used by the scientific community to expedite the calculation and interpretation of next-generation polygenic scores. The development of this platform at its integration with the PGS catalog (Polygenic Score Catalogue) has been developed and it is currently in advanced testing.
Following the project launch during the global challenge posed by the coronavirus pandemic, we have taken active measures to strengthen connections and collaborations across INTERVENE through in-person meetings, workshops, and to strengthen dissemination of the project, for example at international conferences. INTERVENE project postdocs and PIs have also been presenting the project and its results at the leading genetics conferences including American Society of Human Genetics (ASHG) and European Society of Human Genetics (ESHG) conferences. At the ASHG, INTERVENE was featured in a high-profile promotional video of the project in its Thought Leadership Series. INTERVENE partners have also individually and in partnerships with similar initiatives actively participated in promoting the translation of PRS into health care, for example through our ongoing clinical studies, in parallel with developing consensus guideline for trustworthy and deployable artificial intelligence in healthcare, and describing the gaps in knowledge for PRS translation.
Overall, we are moving along our projected track for reaching our objectives and to help realize the next generation of polygenic risk scores for the benefit of science and society.
Application of sophisticated tools and methodologies to the vast collections of biological and health data in biobanks can help us understand how our genetic makeup, together with lifestyle choices and environmental factors can make us prone or resistant to the development of disease.
Over the coming years, INTERVENE will spearhead the development of risk scores that integrate genomic information, health records, and other data types to drastically increase the predictive value, interpretability, and diagnostic capacity of risk scores over the current state-of-the-art.
The harmonization of data from over 1.7 million individuals in INTERVENE’s network of biobanks will, for the first time, provide a critical sample size to powerfully train these scores. It will also extend the predictive ability to include underexplored dimensions of risk prediction, such as risk over time, risk subgroups (such as those dependent on ethic or social status) or overall disease prognosis. INTERVENE’s open-source IGS4EU platform will set new standard methods for integrative genetic prediction and allow other scientists to continuously refine and perfect risk scores across multiple diseases for societal benefit.
The ultimate goal of INTERVENE is harness the power of AI and large-scale health data to enable clinical decision making and better treatment options informed by the genetic makeup of patients, thereby relieving the economic pressure on health care systems, and empowering persons at risk for developing a disease to proactively take measures to reduce the risk of getting sick.
Over the coming years, INTERVENE will spearhead the development of risk scores that integrate genomic information, health records, and other data types to drastically increase the predictive value, interpretability, and diagnostic capacity of risk scores over the current state-of-the-art.
The harmonization of data from over 1.7 million individuals in INTERVENE’s network of biobanks will, for the first time, provide a critical sample size to powerfully train these scores. It will also extend the predictive ability to include underexplored dimensions of risk prediction, such as risk over time, risk subgroups (such as those dependent on ethic or social status) or overall disease prognosis. INTERVENE’s open-source IGS4EU platform will set new standard methods for integrative genetic prediction and allow other scientists to continuously refine and perfect risk scores across multiple diseases for societal benefit.
The ultimate goal of INTERVENE is harness the power of AI and large-scale health data to enable clinical decision making and better treatment options informed by the genetic makeup of patients, thereby relieving the economic pressure on health care systems, and empowering persons at risk for developing a disease to proactively take measures to reduce the risk of getting sick.