Enabling Molecular Breeding

Among the major global food crops, potato is unique for being autotetraploid and clonally propagated. Autotetraploid genomes are organized in groups of four homologous chromosomes, which gives rise to more complex patterns of inheritance than in diploid crops. One focus of the lab is developing new methods and practical software solutions to enable molecular breeding in potato and other autotetraploids, such as alfalfa and blueberry. In 2016 we released GWASpoly, an R package to conduct genome-wide association studies with allele dosage markers. In 2017 we released ClusterCall, an R package for making accurate allele dosage calls from SNP array data.

Rosyara et al. (2016) Software for genome-wide association studies in autopolyploids and its application to potato

Schmitz Carley et al. (2017) Automated tetraploid genotype calling by hierarchical clustering

Endelman et al. (2017) Pedigree reconstruction with genome-wide markers in potato

Genomic Selection

Genomic selection aims to predict the performance of new breeding lines based on phenotypes from related germplasm and genome-wide markers. Since 2014, the Endelman lab has been leading the development of a national training population for potato, with an initial focus on the chip processing market. Breeding value predictions were first used in the UW program in 2018 for yield, specific gravity, and fry color. Ongoing research is extending genomic selection to other market types and traits.

sidebar_fold Created with Sketch.

Endelman et al. (2014) Optimal design of preliminary yield trials with genome-wide markers.

Endelman et al. (2018) Genetic variance partitioning and genome-wide prediction with allele dosage information in autotetraploid potato.

Schmitz Carley et al. (2019) Genetic covariance of environments in the Potato National Chip Processing Trial.


One of the main drivers of genomic selection accuracy is the size of the population used to train the prediction model. At present, the cost of the potato SNP array is more limiting than the cost of phenotyping for increasing our training population size. One way to reduce genotyping costs is through multiplexing, in which many DNA samples are pooled into one library for sequencing. This genotyping-by-sequencing (GBS) approach is widely used in diploids but is less straightforward in tetraploids because of the high read depth needed to determine allele dosage. Research is underway on the use of GBS in potato.

Genetics of tuber skin set and appearance

The appearance and integrity of potato tuber skin has a major influence on marketability and disease resistance during storage, but there is much we do not understand about the genetics of these traits. We are currently using image analysis to quantify skin set and red color in several mapping populations.

Endelman and Jansky (2016) Genetic mapping with an inbred line-derived F2 population in potato