Scientists have long relied on statistical models to predict disease risks and uncover genetic associations, yet many of these tools function as “black boxes,” delivering results without providing a clear view of how individual genetic factors contribute to disease. In a new study published in PNAS, physicist Natália Ružičková, Ph.D. student at the Institute of Science and Technology Austria (ISTA), along with colleagues Michal Hledík and Professor Gašper Tkačik, has developed a model that might offer fresh perspectives on polygenic diseases, where many genomic regions collectively impact health outcomes. Their approach combines genome-wide data with a mathematically formulated model, allowing scientists to analyze and interpret the ways specific mutations contribute to complex diseases.
Unpacking Polygenic Disease Complexity
Polygenic diseases, such as type 2 diabetes, result not from a single genetic mutation but from the combined influence of hundreds or even thousands of variants. Many of these mutations reside in regions with unknown biological functions, yet their cumulative impact on disease susceptibility is significant. “Sometimes, there are hundreds or thousands of mutations linked to a specific disease,” explains Ružičková. “It was a surprising revelation and conflicted with the understanding of biology we had.”
Genome-wide association studies (GWAS) have been instrumental in identifying genetic variants associated with traits and diseases. This method compares the DNA of individuals with and without a given disease, isolating variants that occur more frequently in the affected group. While GWAS has illuminated connections between specific mutations and disease risk, it often falls short in explaining the biological mechanisms behind these links. The new model developed by Ružičková and her colleagues aims to provide a more interpretable alternative by mapping how mutations affect gene expression and interact within broader regulatory networks.
The Omnigenic Model and Its Quantitative Successor
In 2017, researchers from Stanford University introduced the “omnigenic model,” proposing that cells’ extensive regulatory networks connect genes with diverse functions, making many genes indirectly contribute to disease. According to this model, even mutations in genes without direct links to a disease may influence disease development through their effects on other genes within these networks. As Ružičková explains, “Since genes are interconnected, a mutation in one gene can impact other ones, as the mutational effect spreads through the regulatory network.”
Building on this concept, Ružičková’s team has devised the “quantitative omnigenic model” (QOM), a mathematical formalization that attempts to make the omnigenic model’s principles testable. Their approach applies statistical analysis to a known biological system, providing a more transparent view of how mutations propagate through regulatory networks to impact gene expression and, consequently, disease risk.
A Yeast Model with Implications for Human Genomics
To validate their model, the researchers selected Saccharomyces cerevisiae, commonly known as baker’s yeast. This single-celled eukaryote has a cell structure that mirrors those in more complex organisms, providing a practical testbed for studying genetic regulatory networks. “In yeast, we have a fairly good understanding of how regulatory networks that interconnect genes are structured,” says Ružičková.
By applying the quantitative omnigenic model to yeast, the researchers were able to predict gene expression levels and track how mutations impacted the regulatory network. The model effectively pinpointed genes likely involved in particular traits and identified mutations that contributed to specific outcomes. This ability to model causal mechanisms presents a distinct advantage over GWAS, which typically identifies correlations without revealing the underlying biological processes.
Toward More Interpretable Genetic Models
The team’s goal was not to surpass GWAS in predictive power but to offer a model that provides a clearer causal path from mutation to disease. Traditional GWAS methods act as statistical tools that indicate associations between mutations and diseases without revealing how mutations might contribute to disease at a biological level. In contrast, the new model identifies a chain of events within regulatory networks, offering insights into how mutations can propagate through these networks to influence gene expression and contribute to disease.
Although this model is still far from direct clinical application, it holds promise for expanding our understanding of polygenic diseases. “If you have enough knowledge about the regulatory networks, you could build similar models for other organisms as well,” Ružičková notes. Starting with yeast, this proof of principle suggests potential extensions to human genetics, where regulatory complexity in polygenic diseases could be better understood and potentially addressed in therapeutic contexts.
Original Publication
Ružičková, N., Hledík, M., & Tkačik, G. (2024). Quantitative omnigenic model discovers interpretable genome-wide associations. Proceedings of the National Academy of Sciences, 121(44), e2402340121. https://doi.org/10.1073/pnas.24023401
Unpacking Polygenic Disease Complexity
Polygenic diseases, such as type 2 diabetes, result not from a single genetic mutation but from the combined influence of hundreds or even thousands of variants. Many of these mutations reside in regions with unknown biological functions, yet their cumulative impact on disease susceptibility is significant. “Sometimes, there are hundreds or thousands of mutations linked to a specific disease,” explains Ružičková. “It was a surprising revelation and conflicted with the understanding of biology we had.”
Genome-wide association studies (GWAS) have been instrumental in identifying genetic variants associated with traits and diseases. This method compares the DNA of individuals with and without a given disease, isolating variants that occur more frequently in the affected group. While GWAS has illuminated connections between specific mutations and disease risk, it often falls short in explaining the biological mechanisms behind these links. The new model developed by Ružičková and her colleagues aims to provide a more interpretable alternative by mapping how mutations affect gene expression and interact within broader regulatory networks.
The Omnigenic Model and Its Quantitative Successor
In 2017, researchers from Stanford University introduced the “omnigenic model,” proposing that cells’ extensive regulatory networks connect genes with diverse functions, making many genes indirectly contribute to disease. According to this model, even mutations in genes without direct links to a disease may influence disease development through their effects on other genes within these networks. As Ružičková explains, “Since genes are interconnected, a mutation in one gene can impact other ones, as the mutational effect spreads through the regulatory network.”
Building on this concept, Ružičková’s team has devised the “quantitative omnigenic model” (QOM), a mathematical formalization that attempts to make the omnigenic model’s principles testable. Their approach applies statistical analysis to a known biological system, providing a more transparent view of how mutations propagate through regulatory networks to impact gene expression and, consequently, disease risk.
A Yeast Model with Implications for Human Genomics
To validate their model, the researchers selected Saccharomyces cerevisiae, commonly known as baker’s yeast. This single-celled eukaryote has a cell structure that mirrors those in more complex organisms, providing a practical testbed for studying genetic regulatory networks. “In yeast, we have a fairly good understanding of how regulatory networks that interconnect genes are structured,” says Ružičková.
By applying the quantitative omnigenic model to yeast, the researchers were able to predict gene expression levels and track how mutations impacted the regulatory network. The model effectively pinpointed genes likely involved in particular traits and identified mutations that contributed to specific outcomes. This ability to model causal mechanisms presents a distinct advantage over GWAS, which typically identifies correlations without revealing the underlying biological processes.
Toward More Interpretable Genetic Models
The team’s goal was not to surpass GWAS in predictive power but to offer a model that provides a clearer causal path from mutation to disease. Traditional GWAS methods act as statistical tools that indicate associations between mutations and diseases without revealing how mutations might contribute to disease at a biological level. In contrast, the new model identifies a chain of events within regulatory networks, offering insights into how mutations can propagate through these networks to influence gene expression and contribute to disease.
Although this model is still far from direct clinical application, it holds promise for expanding our understanding of polygenic diseases. “If you have enough knowledge about the regulatory networks, you could build similar models for other organisms as well,” Ružičková notes. Starting with yeast, this proof of principle suggests potential extensions to human genetics, where regulatory complexity in polygenic diseases could be better understood and potentially addressed in therapeutic contexts.
Original Publication
Ružičková, N., Hledík, M., & Tkačik, G. (2024). Quantitative omnigenic model discovers interpretable genome-wide associations. Proceedings of the National Academy of Sciences, 121(44), e2402340121. https://doi.org/10.1073/pnas.24023401