Mapping and QTL analysis in polyploid crops

Mapping and QTL analysis in polyploid crops

Many important ornamental, vegetable and field crops are polyploids. Genetic analysis of these crop is still challenging due to their complex nature.

In order to increase progress in research and breeding of this important set of polyploid crops, it is highly desirable to make available genotyping tools, methodology and software for genetic analysis.

The characteristics of polyploid crops is that they have not just two copies per chromosome, but for example four (tetraploids), six (hexaploids) or eight (octoploids). Examples of such crops are cut rose and garden rose, Pelargonium, Petunia, Hippeastrum, Freesia, Alstroemeria, Gladiolus, leek, potato, alfalfa (tetraploid), wheat (tetraploid or hexaploid), chrysanthemum (hexaploid) and strawberry (octoploid).

Bottleneck in genetic analysis of polyploid crops

  • multiple alleles at marker loci and at loci for traits of interest
  • different possibilities of allele dosages across the homologous chromosomes (nulliplex, simplex, duplex, triplex, quadruplex for a tetraploid)
  • dominant markers, based on presence/absence, hide the underlying dosage of the marker allele, and because of the larger number of possibilities in polyploids this leads to large amounts of ‘unobserved’ or incomplete marker data.
  • occurrence of both bivalent and multivalent pairing during meiosis in autopolyploids
  • possibility of preferential pairing of different homologs
  • possibility of different recombination frequencies for different pairs of homologs
  • the higher number of possibilities for the linkage phases between markers
  • no clear distinction between cosegregation due to linkage on the same homologous chromosome or cosegregation due to the presence of alleles on two different homologs that end up in the same progeny individual
  • existence of meiotic phenomena such as double reduction giving rise to progeny genotypes that would otherwise be impossible (AAAa x AAAa can result in aaaa progeny).
  • many polyploid crops are cross pollinators and do not allow selfing (rose, potato) and homozygous parents are usually not available.
  • good quality sequence information for many polyploids is often not (yet) available

As a consequence, genetic and genomic tools that are now standard for diploid crops, are hardly available for polyploid crops.

In order to increase progress in research and breeding of this important set of crops, it is highly desirable to make available genotyping tools, methodology and software for genetic analysis. The recent developments in high-throughput genotyping platforms such as SNP arrays, combined with possibilities for dosage scoring of SNPs in polyploid crops (Voorrips et al., 2011), and higher computational power, now allow the development of these tools, where this was not very well possible before.

Therefore we are currently focusing on the development of a strategy, methodology and software for the construction of polyploid linkage maps, QTL analysis and haplotyping in a number of polyploid crops. From our research at Wageningen UR Plant Breeding and in national and international collaborations with breeding companies, biotechnology companies, and other universities or institutes we have designed SNP scoring platforms for potato, cut rose and garden rose, leek, strawberry and aim to do so also in new collaborations involving other ornamental crops such as Alstroemeria or lily, chrysanthemum and additional polyploid fruit crops.

For example in potato, genotyping-by-sequencing of 83 cultivars helped to identify almost 130,000 sequence variants in 2.1 Mb DNA sequence from 800 loci. A subset of these were used to design a 20K Infinium SNP array which was used for genotyping a large progeny of a tetraploid potato cross and a wide panel of potato germplasm for which also phenotyping data was available.

For cut rose and garden rose a SNP discovery panel of different cultivars was used and over 100 K reliable SNPs were identified using a number of quality criteria. In collaboration with the Leibnitz University Hannover and Affymetrix Inc, we are developing an Axiom® genotyping array for tetraploid rose with over 60,000 potential SNP markers, which will be scored on mapping populations of garden rose and cut rose, a panel consisting of tetraploid cut rose and garden rose cultivars and other rose accessions varying in ploidy level from diploid to pentaploid (5x). For the mapping populations we have collected and still are collecting and analyzing phenotype data for important traits such as disease resistance, frost tolerance (garden roses), and flower colour.

Progeny from single crosses as well as pedigreed populations and collections of breeding material have been genotyped and are being analyzed, using recently developed software tools such as fitTetra (Voorrips et al., 2011) for dosage scoring of SNP markers, pedigreeSim (Voorrips and Maliepaard, 2012) for simulating data from polyploid segregating populations and we are developing additional scripts for estimation of pairwise recombination frequencies among SNP markers, for visualization of SNP haplotypes and for QTL analysis using haplotype information. Marker-trait association analyses have been performed for the potato germplasm panel data and the relationship between allele dosage and the trait values of quantitative traits have been investigated.

These methods and tools enable us to construct polyploid linkage maps of these crops, identification of multi-SNP haplotypes in progenies and breeding material, which can be used in QTL analysis to find marker haplotypes associated with important phenotypic traits such as disease resistances and quality traits and which we hope will, in the future, enable haplotype-assisted selection and haplotype-informed choice of crossing parents for these crops.

Acknowledgements: parts of this work have been co-funded by TTI-Green Genetics, USDA-RosBreed I, the Centre for Biosystems Genomics and breeding companies.