Breeding Differently—the Digital Revolution: High-Throughput Phenotyping and Genotyping.

A conventional potato breeding strategy uses targeted outcrossing, followed by phenotypic recurrent selection over a series of generations to identify improved cultivars. This paper reviews recent research in Australia aimed at improving the efficiency of such breeding. To develop marker-assisted selection (MAS) for traits of interest, our initial targets were qualitative disease resistances for potato cyst nematode (Globodera rostochiensis Ro1), Potato virus Y and Potato virus X. We undertook a cost analysis comparison between MAS and conventional screening, confirming that MAS would be cost-effective within a breeding programme. Then, as the majority of target traits are quantitative in nature, we also looked at methods to address these traits, including progeny testing and a quantitative genetic analysis technique to develop estimated breeding values (EBVs). We found the markers were useful for detecting the disease resistance characters, while the EBVs improved the analysis of the complex traits. Using a combination of MAS, EBVs and conventional screening methods, we then designed a breeding scheme for rapid selection of cultivars with multiple desirable traits, reducing the breeding cycle from over 10 to 4 years. We then explored the factors that will affect the application of genomic selection in potato and investigated strategies to incorporate genomic selection in potato breeding, as we found that it would accelerate genetic gain as the breeding cycle can be reduced to 1 year. Improvements in computational power are also flowing on to research capabilities such as sequencing, high-throughput phenotyping and data analysis, which will accelerate germplasm improvement and breeding. High-throughput phenotyping facilities are being developed that include automated glasshouse systems equipped with imaging sensors and in-field high-throughput phenotyping systems with sensors mounted on ground- or aerial-based vehicles. Using these technological improvements in phenotypic and genotypic analysis will reduce the breeding cycle in a cost-effective manner and means that we can now breed differently. [ABSTRACT FROM AUTHOR]