Structural Variation
Each genome contains unique information.
While traditional genomic approaches focus on small DNA differences, a significant part of genetic variation is found in larger structural changes.
In Work Package 3, we focus on identifying and using this structural variation to create a more complete understanding of the genome. This provides a broader foundation for analysing genetic differences and supports more accurate interpretation of complex traits.
→ go back to all work packages
Why it matters
Genomic selection has significantly improved breeding, but it does not yet capture the full complexity of genetic variation.
A substantial part of this variation lies in structural differences within the genome, which are not fully represented in commonly used approaches.
As a result, important biological signals may remain hidden, limiting the ability to fully explain differences between animals and to understand the genetic basis of complex traits.
Without a more complete representation of the genome, there is a risk that breeding decisions are based on incomplete information.
What we do
Within Work Package 3, we focus on making optimal use of structural variation in breeding.
Each genome contains unique information. While traditional genomic approaches mainly focus on small DNA differences, a substantial part of genetic variation is found in larger structural changes. These include insertions, deletions and rearrangements, which can have a direct impact on observable traits.
To capture this variation, we develop high quality genome resources for livestock populations. This includes building pan-genomes that combine information from multiple individuals and better represent the diversity present in real populations. By moving beyond a single reference genome, we create a more complete view of the genetic landscape.
At the same time, we develop methods and pipelines to detect and use structural variants in practice. This includes improving existing tools, such as genome profiling approaches, and developing new approaches for identifying and imputing structural variation in large datasets.
Work Package 3 also works closely with Work Package 4 to integrate this additional genetic information into genomic prediction models. In doing so, we evaluate how structural variation contributes to phenotypes and how it can improve prediction accuracy.
In addition, we explore emerging approaches such as genomic language models to better prioritise and interpret genetic variation, opening new directions for the future of breeding.
Impact
Work Package 3 expands the genetic foundation of breeding programs by unlocking information that is not captured by traditional approaches.
By incorporating structural variation into breeding, it becomes possible to better explain differences between animals and improve the accuracy of genomic predictions. This allows breeding organisations to make more informed selection decisions and to make better use of the available genetic diversity.
The development of pan-genomes and practical pipelines ensures that these innovations are not only theoretical, but can also be applied in real breeding programmes. This bridges the gap between advanced genomic research and day to day breeding practice.
Ultimately, Work Package 3 enables a more complete and realistic representation of the genome, supporting more precise, data driven and future ready breeding strategies.