The Challenge: Potato is a genetically complex autotetraploid. Standard sequencing collapses its four chromosome copies into one, losing critical variation.

Our Work: We constructed a haplotype-resolved pan-genome of European potato. This revealed extraordinarily high sequence diversity between haplotypes (20x higher than in humans) but surprisingly low haplotype diversity across the continent due to historical breeding bottlenecks.

While nature offers thousands of edible plant species, global agriculture is built around a small set of widely cultivated staple crops. We are working to change that. By harnessing the power of modern genomics, we aim to unlock this untapped potential and revive forgotten crops through de novo domestication.

Meet the Aardaker (L. tuberosus).

Native to Eurasia, this member of the Fabaceae family is a nitrogen-fixing powerhouse that produces protein-rich tubers. Our work is driven by fundamental scientific curiosity, coupled with a clear translational goal. We are generating a chromosome-scale reference genome and conducting a holistic population genomics analysis to uncover the species’ evolutionary origins, characterize its existing genetic architecture, and assess its adaptive potential.

We work directly with breeders and openly share our insights to ensure the Aardaker makes the leap from scientific curiosity to a resilient, stable crop.

More on the applied work can be found on the Aardaia webpage →

Aardaker field

Field trial

Field plot

Growing in the field

Selected Tubers from single plant

Aardaker pollinator

Harvested tubers

The protein potato

Root nodules for nitrogen fixation

Young aardaker plant

Large aardaker plant in the greenhouse

The Mystery of "Sport" Mutations: Many fruit varieties, such as the dark-red 'RubyMac' apple, arise spontaneously as somatic mutations on a single branch of a tree. Understanding these "bud sports" is crucial for breeding improved cultivars with desirable traits.

Genomic Forensics: We sequenced the genome of a single apple tree to identify the specific mutations separating wild-type and mutant branches. We discovered that somatic evolution is driven not just by new mutations, but frequently by gene conversions (loss of heterozygosity) that are specific to individual cell layers in the meristem.

A Versatile Platform for Haploid Genetics: We are pioneering single-nucleus sequencing technologies to study the haploid life cycle of plants. By isolating and sequencing individual pollen nuclei, we can treat each gamete as an independent "offspring." This allows us to perform high-resolution genetic mapping without the need for time-consuming crosses or large progeny populations.

Beyond Recombination: While our primary focus has been mapping crossover landscapes, this technology is broadly applicable. We are developing methods to analyze gene expression dynamics during pollen development, characterize rare cell types, and investigate the transcriptomic consequences of haploid selection. Our group actively maintains and improves the tools (e.g., coelsch) required to analyze these complex, sparse molecular datasets.

Anthers of the parental lines

Pollen nuclei, Hoechst stained

Deconvolved confocal section

Population-Scale Mapping: We investigate how genetic diversity shapes the landscape of meiotic recombination. By extending our single-cell framework to natural populations, we can identify the specific genetic modifiers, like trans-acting factors or local structural variants like inversions, that control where and how frequently crossovers occur.

Genetics of the Haploid Phase: Beyond recombination, we map the regulatory networks of the gamete itself. Using single-nucleus eQTL mapping, we uncover how cis- and trans-regulatory variation drives gene expression in pollen. For example, we recently identified DUO3 as a master regulator of sperm cell development by linking haplotype inheritance to transcriptomes in single nuclei.

Mapping drivers of interspecific variation in recombination landscape: Species of the Brassicaceae flowering plant family vary in genome structure and ecological characteristics, while their core meiotic genes are well-conserved. We aim to identify how this variation drives differences in recombination landscape between said species and the degree to which specific factors (mating system, ploidy level, haploid chromosome number, chromosomal translocations) contribute.

Deciphering Centromere Dynamics: Centromeres are essential for cell division but evolve rapidly. We use replicated, error-free genome assemblies to track the "birth" of mutations in these repetitive regions. Our work reveals a unique mutational spectrum dominated by non-allelic gene conversions and small, structure-preserving indels.

From Chaos to Order: We model how these fundamental processes shape genome architecture. Our simulations demonstrate that simple, small-scale mutations are sufficient to drive the self-organization of complex, megabase-scale Higher-Order Repeat (HOR) structures, providing a unified theory for centromere evolution.

Reference Bias: A single linear reference genome captures only a fraction of a species' genetic diversity. It introduces reference bias, causing us to miss genes and structural variants that are present in the population but missing from the reference.

Aggregating Diversity: Our group develops pan-genomic approaches to comprehensively characterize and aggregate variation across plant populations and species. By combining long-read sequencing with advanced phasing methods, we construct haplotype-resolved assemblies that reveal the full spectrum of genetic variation.

Evolution of Gene Synteny: Conservation or loss of gene synteny between species can affect gene expression, and subsequently phenotype. Using a panel of diverse Brassicaceae genomes, we investigate conservation of gene synteny across 20 million years of evolution and its implications for gene function.

From chromosome-level pan-genomes of Arabidopsis to phased assemblies of tetraploid potato and comparative resources across the Brassicaceae, our work connects natural variation to phenotypic diversity, providing foundational knowledge for genome evolution, population history, and plant breeding.

We develop methods that leverage genomic differences from SNPs to large chromosomal rearrangements. Our most popular tools include:

• SyRI: Synteny and Rearrangement Identifier. Finds inversions, translocations, and duplications.
• plotsr: A tool to visualize structural similarities and rearrangements between multiple genomes.
• coelsch: A suite of tools for identifying meiotic recombination events from single-nucleus sequencing data.
• findGSE: Estimates genome size by fitting k-mer frequency distributions.
• SHOREmap: Mapping-by-sequencing analysis for forward genetic screens.

All tools, repositories and documentation →

A core part of our work is to continuously evaluate and implement emerging sequencing and genomic technologies, ensuring that our datasets push the boundaries of genome assembly.

Gamete binning: By sequencing large numbers of single gamete genomes, we can separate the two haplotypes of a heterozygous individual before assembly, turning a hard phasing problem into two straightforward ones. The approach now underpins much of our haplotype-resolved assembly work.

All resources, such as assemblies, annotations, and analytical pipelines, are shared openly, supporting the broader community in trait discovery, evolutionary genomics, and plant breeding innovation.