Shuqin Jiang, Qian Cheng, Jun Yan, Ran Fu & Xiangfeng Wang

Theoretical and Applied Genetics May 2020; 133:1491–1502

Key message

We propose a new model to improve maize breeding that incorporates doubled haploid production, genomic selection, and genome optimization.

Abstract

Breeding 4.0 has been considered the next era of plant breeding. It is clear that the Breeding 4.0 era for maize will feature the integration of multi-disciplinary technologies including genomics and phenomics, gene editing and synthetic biology, and Big Data and artificial intelligence. The breeding approach of passively selecting ideal genotypes from designated genetic pools must soon evolve to virtual design of optimized genomes by pyramiding superior alleles using computational simulation. An optimized genome expressing optimal phenotypes, which may never actually be created, can function as a blueprint for breeding programs to use minimal materials and hybridizations to achieve maximum genetic gain. We propose a new breeding pipeline, “genomic design breeding,” that incorporates doubled haploid production, genomic selection, and genome optimization and is facilitated by different scales of trait predictions and decision-making models. Successful implementation of the proposed model will facilitate the evolution of maize breeding from “art” to “science” and eventually to “intelligence,” in the Breeding 4.0 era.

See https://link.springer.com/article/10.1007/s00122-019-03493-z

Figure 5: Combining DH breeding, GS, and genome optimization into a pipeline for improvement of maize breeding. The details of the proposed maize breeding pipeline are described in the main text