Quantitative aspects of Plant Breeding

The group focuses on research aimed at understanding the interaction between genotype, environment and management by using all kinds of different techniques and databases. Therefore database development, maintenance and use is an important area of attention. Furthermore statistical aspects of using large datasets especially in the field of transcriptomics and metabolomics is a major focus area.

Crops: potato, brassica, tomato and other crops of importance.

Plant breeding has a strong link with quantitative genetics, statistics and bioinformatics, for example in identifying regions on crop genomes that are associated with observed variation in phenotypic traits, and in identifying for these genomic regions lists of genes that are present there and narrowing down to the more likely candidate genes using gene annotation and external information, from literature and databases, on gene function, homology across species, genetic and metabolic pathways. There are also many quantitative aspects in analysis and handling of genotype x environment interaction, and in improving results and efficiency of selection. Developments in high-throughput analyses of phenotypic traits, genotype in terms of DNA sequence data and molecular markers, RNA transcripts, proteins and primary and secondary metabolites all provide new opportunities for better prediction of plant phenotypes and discovery of the features (genes, proteins, metabolites) underlying genetic variation.

Research focus

We focus on research and education in the fields of quantitative genetics and genomics, selection procedures, data management and development of genetics tools, statistical and database and bioinformatics tools for analysis and visualization of large data sets. Current activities include combined data analysis of molecular markers, gene expression and metabolomics data and phenotype (e.g. disease resistance or product quality) scored on segregating populations of crosses. Methods being used are procedures such as random forest for classification or multiple regression in cases where the number of predictor variables (e.g. molecular markers, genes, metabolites) is much larger than the number of samples in which they have been measured (plants, tissues). Other areas of interest are modelling genotype x environment interaction, mapping and QTL analysis in single segregating populations, multiple populations or collections of germplasm. A specific focus area is also the development of a genetic analysis pipeline for polyploid crops.

Activities

  • BreeDB
  • Construction of linkage maps and QTL analysis
  • Linkage Disequillibrium mapping and association mapping
  • Epistatis mapping
  • Genetical genomics
  • Genomic prediction and selection
  • Estimation of Genotype x Environment interaction
  • Relating henotype data to microarray, metabolomics or proteomics data
  • Methods and software for genetic analysis in polyploid species

Data Management & Data Integration

Plant breeding has a strong link with quantitative genetics, statistics and bioinformatics, for example in identifying regions on crop genomes that are associated with observed variation in phenotypic traits, and in identifying candidate genes for these genomic regions.

Multi disiplinary

Research within Plant breeding is multi-disciplinary and is dealing with a lot of different types of data, obtained from field-trials but also from high-throughput analysis of molecular markers, RNA transcripts (microarrays), proteins and secondary metabolites.

Data management

An efficient database will help in elucidating the genetics of economically important traits, in identifying molecular markers associated with agronomic traits, in allele mining and choosing interesting accessions fur further breeding with improved traits important for consumers, processors and producers.

Semantic Web

One goal within plant breeding is to find the causal gene(s) explaining a given phenotype. Semantic web technology brings opportunities to integration data and information across spread data sources. For example, Annotex and Marker2sequence are two applications developed by Plant Breeding, which rely on this semantic web technology to integration genes, proteins, metabolites, pathways, and literature.

Contact: Richard Finkers

Statistical approaches

Current activities include combined data analysis of molecular markers, gene expression and metabolomics data and phenotype (e.g. disease resistance or product quality) scored on segregating populations of crosses. Methods being used are procedures such as random forest for classification or multiple regression in cases where the number of predictor variables (e.g. molecular markers, genes, metabolites) is much larger than the number of samples in which they have been measured (plants, tissues). Other areas of interest are modelling genotype x environment interaction, mapping and QTL analysis in single segregating populations, multiple populations or collections of germplasm. A specific focus area is also the development of a genetic analysis pipeline for polyploid crops.

Contact: Chris Maliepaard