A New Algorithm for Genetic Risk Scoring: Closing Disparities in Health Care
Introduction
Genetic risk scoring algorithms have emerged as a promising tool in identifying high-risk populations that could benefit from preventive interventions for various diseases and conditions. These algorithms are based on large-scale genetic studies that connect certain DNA variants to higher or lower disease risks. However, the majority of these studies have focused on people of European ancestry, leading to disparities in health care for populations of non-European ancestries.
The Problem with Current Genetic Risk Scoring Algorithms
Due to genetic differences between populations, the existing risk scoring algorithms developed based on European ancestry data have not always worked effectively in other populations. This has resulted in a performance gap in risk scoring for populations of non-European ancestry, contributing to health care disparities. Many risk scoring models derived from genetic studies in populations of non-European ancestry often fall short because these studies are relatively small in scale.
The Development of a New Algorithm: CT-SLEB
A team led by researchers at the Johns Hopkins Bloomberg School of Public Health and the National Cancer Institute has developed a new algorithm called CT-SLEB (Combining Techniques for Scoring Loci Effects across Broad Ancestry) to address the disparities in health care caused by current risk scoring algorithms.
The CT-SLEB algorithm utilizes a combination of artificial intelligence techniques, including machine learning and Bayesian statistical modeling. It has been trained using data from diverse populations, including data from the 23andMe database, the Global Lipid Genetics Consortium, the National Institutes of Health’s All of Us research program, and the UK Biobank.
The algorithm generates genetic scores for 13 different traits, including health conditions such as coronary artery disease and depression, in five major ancestry categories: European, African, Latino, East Asian, and South Asian. This comprehensive approach aims to improve risk scoring accuracy across diverse populations.
The Performance of CT-SLEB
The comparative analysis conducted by the research team showed that the new ancestry-specific risk scoring models generated by CT-SLEB generally outperformed the standard polygenic risk scoring models, which are primarily based on European ancestry data sets or smaller non-European ancestry data sets. The accuracy of genetic risk scores in populations of African descent, where performance was often the lowest, was particularly improved by CT-SLEB.
In addition to its improved accuracy, CT-SLEB has proven to be computationally faster compared to its competitors. This enables the analysis of a larger number of DNA variants and more populations, making it a more efficient tool for genetic risk scoring.
Limited by Data: The Need for Larger Studies
Although CT-SLEB shows promise in reducing health care disparities, the study concluded that closing the performance gap between European and non-European populations requires larger genome-wide association studies in non-European ancestry populations. While advanced algorithms and artificial intelligence can enhance risk scoring accuracy, the availability of large, well-designed studies is crucial in developing effective algorithms.
Nilanjan Chatterjee, the senior author of the study and Bloomberg Distinguished Professor in the Bloomberg School’s Department of Biostatistics, emphasizes the importance of combining advanced methods with comprehensive data. Machine learning and AI can enhance risk scoring algorithms, but without extensive studies, their effectiveness will be limited.
The Future of Genetic Risk Scoring
The researchers behind CT-SLEB are currently working on more advanced methods that aim to further improve risk scoring accuracy while maintaining computational efficiency. By incorporating cutting-edge technologies and expanding the availability of large-scale studies, they hope to develop risk scoring models that work equally well in non-European ancestry and European ancestry populations.
This research not only has significant implications for reducing health care disparities but also highlights the importance of embracing diversity in genetic studies. By including populations of diverse ancestries, researchers can develop more comprehensive and accurate risk scoring algorithms that benefit all individuals, regardless of their genetic background.
Conclusion
The development of the CT-SLEB algorithm marks an important step forward in reducing disparities in health care caused by genetic risk scoring algorithms. By incorporating data from diverse populations and utilizing advanced techniques, CT-SLEB has shown promising results in improving the accuracy of risk scoring models for non-European ancestry populations.
However, it is crucial to recognize that the full potential of these algorithms can only be realized with larger, well-designed studies. In order to develop risk scoring models that work equally well in populations of both European and non-European ancestries, the availability of comprehensive and diverse genetic data is essential. The ongoing efforts of researchers in this field hold promise for the future of genetic risk scoring and the reduction of health care disparities.
—————————————————-
| Article | Link |
|---|---|
| UK Artful Impressions | Premiere Etsy Store |
| Sponsored Content | View |
| 90’s Rock Band Review | View |
| Ted Lasso’s MacBook Guide | View |
| Nature’s Secret to More Energy | View |
| Ancient Recipe for Weight Loss | View |
| MacBook Air i3 vs i5 | View |
| You Need a VPN in 2023 – Liberty Shield | View |
A team led by researchers at the Johns Hopkins Bloomberg School of Public Health and the National Cancer Institute has developed a new algorithm for scoring genetic risk for major diseases in populations of diverse ancestries that promises to reduce disparities in health care.
Genetic risk scoring algorithms are considered a promising method for identifying high-risk groups of people who could benefit from preventive interventions for various diseases and conditions, such as cancers and heart diseases. These risk scoring algorithms are based on large-scale genetic studies that link certain DNA variants with higher or lower disease risks.
The vast majority of subjects in these genetic studies have been people of European ancestry. The resulting risk scoring algorithms have not always worked well in other populations, due to genetic differences between populations.
The new method, described in an article appearing online today at Nature genetics, has been applied to data from genetic studies from 23andMe Inc. and other sources involving more than 5 million people from diverse populations to generate genetic scores for 13 traits, including health conditions such as coronary artery disease and depression, in five different ancestry categories: European. , African, Latino, East Asian and South Asian. The researchers also tested the new method in large-scale simulation studies.
“We show that our method can help close the performance gap in risk scoring for populations of non-European ancestry,” says the study’s senior author, Nilanjan Chatterjee, PhD, Bloomberg Distinguished Professor in the Bloomberg School’s Department of Biostatistics. . “At the same time, we also concluded that we cannot completely close the gap with new methods alone; we also need larger data sets on these populations.”
Many risk scoring models derived from genetic studies in populations of non-European ancestry often fall short because those studies are typically relatively small in scale. This results in a risk score performance gap between European and non-European ancestry populations, which may contribute to health care disparities.
The new method, which the researchers call CT-SLEB, used a combination of artificial intelligence techniques including machine learning and Bayesian statistical modeling. In addition to the 23andMe database, the researchers “trained” CT-SLEB with data from the Global Lipid Genetics Consortium, the National Institutes of Health’s All of Us research program, and the UK Biobank.
The research team’s comparative analyzes showed that these new ancestry-specific risk scoring models for non-European populations generally outperformed standard polygenic risk scoring models that are based primarily on European ancestry data sets, or on of smaller data of non-European ancestry. .
The researchers also compared CT-SLEB with several alternative methods. They found that the proposed method is particularly useful for improving genetic risk scores in populations of African descent, where scoring accuracy is generally the lowest. The team also found that CT-SLEB is computationally much faster compared to its closest competitors and could therefore analyze a much larger number of DNA variants and more populations.
The team is now working on more advanced methods that perform even better but are still computationally fast, Chatterjee says.
It also emphasizes that, as the team’s calculations in the study showed, having polygenic risk scoring models that work equally well in non-European ancestry and European ancestry populations will require more genome-wide association studies in non-European ancestry populations. .
“Many people think that machine learning and AI can do magic, but without large, well-designed studies, algorithms won’t be as useful,” says Chatterjee.
The lead author of the paper is Haoyu Zhang, PhD, who was a doctoral student at the Bloomberg School at the time the study began and is currently a researcher at the National Cancer Institute. 23andMe researchers contributed to the development of the new method and analysis of the data. The CT-SLEB code is publicly available through GitHub. The code availability section of the document includes a link to GitHub that includes the CT-SLEB code.
Funding was provided by the National Institutes of Health (K99 CA256513-01, R00 HG012223, 5T32HL007604-37, R35-CA197449, U19-CA203654, R01-HL163560, U01-HG009088, U01-HG012064, R01 HG0 10480-01 and U01HG011724 ).
—————————————————-