Geneticists have long used constraint modeling to understand disease-causing mutations in the nuclear genome. Constraint models allow researchers to assess selective pressure that causes certain gene variants to be eliminated over time, revealing patterns that help pinpoint disease-associated genetic changes. However, these models have been largely ineffective when applied to the mitochondrial genome, leaving geneticists without a reliable way to identify disease-causing mutations in mitochondrial DNA (mtDNA). This limitation has been frustrating for researchers and families affected by genetic illnesses linked to mitochondrial mutations.
In a significant breakthrough, a team of Yale researchers led by geneticists Nicole Lake and Monkol Lek has developed a new approach that finally addresses this challenge. The team’s novel model, detailed in the journal Nature, offers geneticists an essential tool to identify which mtDNA mutations are likely to contribute to disease. This advancement represents a long-awaited leap forward in mitochondrial research, providing a more comprehensive understanding of the mitochondrial genome and its role in health.
Mitochondria are unique cellular structures responsible for producing energy and regulating processes such as programmed cell death. Unlike nuclear DNA, mitochondrial DNA is inherited solely from the mother and has distinct properties that make it challenging to analyze with standard models. Although many tools exist for studying nuclear genome mutations, similar tools for mtDNA have been lacking due to factors like the small size of the mitochondrial genome and its unique genetic features.
To overcome these barriers, Lake and Lek’s team designed a model based on an entirely new methodology that adapts a “composite likelihood” approach, which was previously used to solve complex issues in other areas of genetics. This approach allowed the researchers to develop a mitochondrial mutational model that analyzes genetic mutations at various locations within the mitochondrial genome, calculating the likelihood of mutations occurring at specific sites. By studying patterns of change across the genome, the model reveals which regions are constrained—meaning they experience fewer changes over time and are likely to be critical to health.
This model builds on prior research by Lake and Lek, who collaborated with colleagues at the Broad Institute of MIT and Harvard to generate a comprehensive mitochondrial dataset. This initial dataset of 56,434 individuals served as the foundation for their current work, allowing the team to identify constrained regions in the mitochondrial genome that are more likely to harbor mutations tied to disease. By quantifying these constrained regions, the researchers could create a detailed map showing where disease-causing mutations are most likely to occur.
To further verify their findings, the Yale team expanded their dataset using the UK Biobank, a vast resource of genetic and health information from around half a million participants. Applying their model to this large sample of individuals with mitochondrial disease, the researchers confirmed that the constraint model effectively identified mtDNA variants that contribute to disease. This success demonstrated that the model could be adapted for use by other researchers worldwide to study mitochondrial genome variations and their links to disease.
The impact of this tool could be profound, especially for addressing “variants of uncertain significance,” or genetic changes in mtDNA whose effects on health are unclear. These variants often cause confusion and anxiety for families dealing with genetic illnesses because they cannot be definitively linked to disease without further evidence. Lake is hopeful that their new tool will reduce this uncertainty, providing geneticists with clearer insights into which variants are harmful.
The development of this model was a collaborative effort involving experts from multiple disciplines. Lake credits a series of fortuitous discussions for helping the team achieve their “eureka” moment. In early 2020, just before the COVID-19 pandemic forced widespread shutdowns, Lek discussed their research challenges with Shamil Sunyaev, a computational geneticist at Harvard, during his visit to Yale. Sunyaev mentioned encountering similar challenges in cancer research and proposed using a composite likelihood model, which ultimately provided the foundation for the mitochondrial constraint model. With the pandemic limiting in-person meetings, the research team turned to virtual collaboration, involving experts such as Dan Arking from Johns Hopkins University to refine the model and validate it with additional datasets.
Through close collaboration and feedback from leading geneticists, Lake and Lek’s team built a mutational model that was not only predictive but also consistently delivered reliable results. This collaborative approach, Lake and Lek believe, was essential to their success. Lake describes their model as a “first-generation tool” that will be freely accessible to scientists around the world, emphasizing that it has “exciting potential for expansion.”
This breakthrough highlights a long-overlooked part of the genome and underscores the importance of mtDNA in health and disease. As research continues, the mitochondrial constraint model is expected to serve as a foundation for further exploration of the mitochondrial genome. This innovative tool could open up new pathways for understanding and potentially treating mitochondrial diseases, providing hope to those affected by these conditions. The model’s future development may lead to a deeper understanding of mtDNA’s influence on human health, and its role in diseases beyond those currently known to have a mitochondrial link.
Source: Yale University