Embedded Files
The intricate patterns of genetic variation we observe today hold within them the echoes of countless ancestral events, shaping the evolutionary path that brought us to this moment. Woven into our genomes are traces of evolution, migrations, separations - alongside the work of molecular processes that copy, shuffle, and alter DNA. These layers of history, from the distant past to the recent present, leave signatures that can be decoded with the right data, theory and algorithms. Our lab develops and applies computational approaches to uncover these signatures and make sense of the variation, and apply those insights to better understand evolution, history, and health.
Population genetics at scale
New sequencing technologies enable the collection of large and diverse genomic datasets. We develop population-genetic computational methods that can analyse this new scale and resolution of data, with a focus on inferring demographic and evolutionary histories at multiple levels - from species-wide patterns to individual-level variation and specific genetic variants.
Questions:
How did populations grow, shrink, and mix over time - and how can we recover these events from the traces they leave in our genomes?
How can we recover the evolutionary history of particular mutations or of more complex traits?
How can modern machine learning techniques improve and complement classical models and algorithms in answering these questions?
Molecular mechanisms of variation
Recombination shuffles haplotypes into mosaics during sperm and egg formation, contributing to genetic diversity. Using accurate long-read sequencing, we study recombination, with a focus on non-crossover recombinations (gene conversions). Investigating their timing, locations, and underlying molecular mechanisms deepens our understanding of recombination and its roles in infertility, various genetic disorders, cancer risk, and population-genetic models.
Questions:
Where along the genome do non-crossover recombinations (gene conversions) occur, and how does their frequency vary between individuals, samples, and species? What determines this variability?
What molecular mechanisms trigger non-crossovers? Do they involve known DNA repair pathways?
Can a better understanding of non-crossovers help explain cancer predisposition or recombination-related disorders like Bloom syndrome?
New models for newly accessible genomic regions
Long-read sequencing only now allows us to access previously unmappable regions of the genome, including telomeres, centromeres, and transposable elements. These regions evolve differently from the rest of the genome, and so could offer new windows into the past. We are interested in developing new theoretical frameworks and methods to explore the evolutionary insights they offer, and to study their role in health and disease.
Questions:
What theoretical frameworks are needed to study the evolution of complex, repeat-rich genomic regions?
How can we reconstruct the history of sequences that duplicate, move, and evolve outside standard models of inheritance?
Google Sites
Report abuse