Our projects connect genome evolution, genetic variation and crop improvement with the development of genomic resources, analytical tools and data infrastructure. We work across academic institutes, international centres, breeding companies and international consortia.

Transversal themes

Understanding hybridisation and introgression:

One of the lab’s long-standing questions is how hybridisation and gene flow reshape genomes and phenotypes. In Vietnamese rice, we investigated population structure, selection and introgression within locally adapted landraces. In common bean, we characterised admixture between the Mesoamerican and Andean gene pools and examined how introgressed genomic regions relate to domestication-associated and stress-response traits. These systems established a research framework that now extends across our programme: combine population history with functional and quantitative genomics rather than treating genetic associations independently of the evolutionary processes that generated them.

Genome evolution in complex crop systems:

In banana and tropical forage grasses, our work has increasingly focused on the genomic consequences of hybridisation and polyploidisation. We develop and apply approaches for detecting ancestry mosaics, structural and copy-number variation, dosage differences, introgression and homoeologous exchange. Long-read sequencing and pangenomic approaches now allow us to characterise these processes at increasingly high resolution. Together, these projects provide the biological and computational foundation for our current work on genomic variation, prediction and community resources for crop improvement.

Projects on this page

Complex genomes, pangenomes and genetic variation

White clover: pangenomics for climate-resilient breeding

With Germinal Horizon and IBERS, we are developing genomic resources for white clover, an agriculturally important but comparatively under-resourced forage crop. The project combines a European diversity panel, environmental information and new long-read sequencing to generate multiple high-quality genome assemblies and develop a white-clover pangenome. We are using this framework to characterise variation associated with cold tolerance, persistence and environmental adaptation, while also testing whether pangenome-based approaches improve variant discovery and association mapping relative to a single reference genome. This project directly links evolutionary genomics, quantitative genetics and practical plant breeding.

Further reading: Unlocking the agricultural value of clover.

DECODE: variation beyond a single reference genome

Within the BBSRC Institute Strategic Programme Decoding Biodiversity (DECODE), we investigate how introgression, chromosomal rearrangements, structural variation and other forms of complex diversity contribute to agriculturally important traits. A central theme is the transition from single reference genomes towards multiple high-quality assemblies, long-read population datasets and richer representations of genomic variation. These approaches are particularly important in hybrid, polyploid and highly diverse crop systems where biologically important variation can be missed by SNP-centred analyses alone. Further reading: Characterising gene function, biosynthetic pathways and variation in agri/aquacultural traits.

Polyploid and hybrid genome assembly

Long-read sequencing has allowed us to address technical and biological questions in complex polyploid genomes. We have completed and published haplotype-resolved chromosome-level assemblies from PacBio and Oxford Nanopore data and continue to develop end-to-end workflows for assembly, quality control and structural-variation analysis. Our work in Urochloa and banana examines ancestry, introgression, dosage, homoeologous exchange and chromosomal imbalance, helping fill methodological gaps in crop systems for which many standard genomics tools remain optimised for diploid organisms.

Genomics for breeding

Celery and vegetable breeding with Tozer Seeds

Through an Innovate UK Knowledge Transfer Partnership with Tozer Seeds, we are embedding genomics and computational approaches directly within a commercial vegetable-breeding programme. The project characterises diversity in breeding material and uses multi-trial association analyses to identify markers that can support marker-assisted selection. It also provides a practical test of how genomic methods transfer between crop systems and how population structure, domestication history and trial design affect the robustness of association mapping. The collaboration links academic genomics with the operational requirements of a breeding company, helping develop tools and approaches that can be sustained beyond an individual research project.

Further reading: Collaborating to create resilient celery.

Common bean: domestication, adaptation and prediction

We study diversity across the Mesoamerican and Andean gene pools of common bean and their admixed populations in Colombia and neighbouring regions. Our work has mapped QTL for domestication-associated traits, including photoperiod sensitivity and determinacy, and has used population genomics to distinguish trait associations from the effects of population structure and introgression. We have also investigated variation in drought-response strategies and markers with potential value for breeding. More recently, this work has expanded towards phenotype prediction, including benchmarking mixed and machine-learning models and examining how genomic prediction can support sparse testing across environments.

Integrated data for breeding

Legume Generation: community genomic resources and data infrastructure

Within the Horizon Europe Legume Generation programme, we co-lead work to develop the project’s digital Knowledge Centre and Legume Discovery platform. The consortium brings together research organisations and breeding programmes across multiple countries and crops, with phenotypic and genetic data contributed by partners working on soybean, pea, lentil, common bean, lupin and clover. Our work is moving towards a single integrated environment for plant-evaluation, multi-environment trial and genotyping data, replacing the initially proposed separation between phenotypic and marker-data platforms.

The platform is being developed around FAIR principles and the practical needs of the project’s crop Innovation Communities. Current work includes:

  • integrating trial and phenotypic datasets from multiple partners;
  • harmonising metadata and developing standardised trait ontologies;
  • creating crop-specific interfaces for data discovery and exploration;
  • supporting multi-environment analysis with mixed-linear models, BLUEs and BLUPs;
  • comparing alternative normalisation models so users can assess their performance;
  • integrating genotyping and marker information;
  • developing the backend required for downstream GWAS and marker-to-target discovery.
  • The technical development is led within the group by a computational PDRA, James Brett, with scientific requirements shaped through regular engagement with the crop Innovation Communities and other project partners in the consortium.

This work reflects a broader aim of the lab: to make genomic and phenotypic information not only available, but findable, interoperable, analytically useful and reusable by researchers and breeders.

Further reading: Boosting beans for breeders and Standout innovation contributes to knowledge exchange.

Crop resilience

Banana diversity and Fusarium resistance

Banana provides a model for understanding the consequences of hybridisation, polyploidy, clonal propagation and somatic variation in a crop of major global importance. We have mapped introgression and homoeologous exchanges across cultivated banana diversity and linked genomic variation to fruit, plant architecture and yield-related traits. Building on this population framework, we established macropropagation and Fusarium infection assays in collaboration with Tropic Biosciences, allowing representative genotypes to be tested for resistance to TR4. We have also developed a banana pan-NLRome and combined NLR presence/absence variation with disease-response phenotypes to prioritise candidate resistance loci. This connects population genomics, structural variation and experimental validation with routes towards precision breeding and genome editing.

Further reading: Bananas are on the brink but close cousins could save their skins.

Tropical forage genomics and insect-pest resistance

Our long-standing work on Brachiaria/Urochloa tropical forage grasses combines genome resources, quantitative genetics and breeding applications. We have generated reference and haplotype-resolved genome assemblies, developed genetic maps and identified markers for agronomically important traits including apomixis. More recently, we have combined high-throughput image phenotyping and genome-wide association analysis to investigate resistance to the hemipteran spittlebug Aeneolamia varia. This work places genomic analysis directly at the crop–insect pest interface, while maintaining a strong connection to forage breeding and sustainable livestock production.

Further reading: Study identifies candidate genes to accelerate tropical forage breeding.

Rice diversity and breeder training

One of our earlier projects investigated the population structure and adaptation of Vietnamese rice landraces in the context of global Asian rice diversity.

Working with local partners, we identified distinctive Vietnamese diversity, including an indica population around the Red River Delta carrying genome-wide japonica introgression, and linked signatures of selection and introgression with candidate traits relevant to adaptation and breeding. The project also supported international training and capacity building in genomics and bioinformatics.

Further reading: Rice crops in a changing climate.

Tropical forage partnerships

Our forage-genomics work has involved long-standing collaboration with CIAT and ILRI, linking genomic resources and breeding tools to crops important for livestock production in tropical regions. These collaborations have combined genome sequencing, quantitative genetics, marker development and training, with the aim of making genomics directly useful to breeding programmes operating in regions where forage improvement can contribute to more productive and sustainable livestock systems.