Breeding of woody perennials is hindered by long juvenile periods, prevailing outcrossing reproductive systems, and complex genomic architectures. In tropical regions, the pace of genetic gains is further constrained by local-scale environmental variation, recurrent G × E, and lack of funding for long-term tree improvement initiatives. Fortunately, indirect selection strategies offer a fast track to sidestep some of these bottlenecks, as we explore throughout this chapter. From a conceptual point of view, age-age correlations and proxy traits constitute the first instances of indirect selection. After all, complex quantitative traits in mature trees are forecasted by relying on measures of well-correlated traits earlier in the trees’ life cycle. However, with the advent of the genomic era, the speed and precision of indirect selection has been accentuated by screening the underlying genetic markers, enabling molecular breeders to trace and select for heritable additive variation rather than earlier trait segregation. At the dawn of the technology, predictions were built upon few associated markers linked with the desirable traits, an approach referred to as marker assisted selection (MAS). Meanwhile, increasing marker sets and sequenced genomes piled up, which made genome-wide predictions feasible as part of the modern genomic selection paradigm. When combined with other “omics” technologies such as transcriptomics, metabolomics, phenomics, and enviromics, the spectrum of inferences scales up, empowering more robust machine learning (ML)-based inferences across entire metabolic pathways, their phenotypic consequences, and the plant breeding triangle (genome + phenome + envirome). As if these progressions were not fascinating enough, tree breeders are starting to implement speed breeding, in which species’ life cycles are shorten even more by a combination of hormones, grafting, light treatments, growth chamber facilities, and gene editing of flowering genes. This way, precocious seedling and saplings can reach flowering maturity only few months into the greenhouse, allowing an expeditious first breeding generation. When coupled together, genomic selection and speed breeding promise quick and precise tree breeding cycles, easing population improvement at the pace imposed by climate change, environmental variation, business turnover, and extramural funding in tropical regions.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Breeding Without Breeding: Enabling Indirect Selection Schemes for Tropical Tree Improvement

  • Santiago Bedoya-Londoño,
  • Gloria P. Cañas-Gutiérrez,
  • Andrés J. Cortés

摘要

Breeding of woody perennials is hindered by long juvenile periods, prevailing outcrossing reproductive systems, and complex genomic architectures. In tropical regions, the pace of genetic gains is further constrained by local-scale environmental variation, recurrent G × E, and lack of funding for long-term tree improvement initiatives. Fortunately, indirect selection strategies offer a fast track to sidestep some of these bottlenecks, as we explore throughout this chapter. From a conceptual point of view, age-age correlations and proxy traits constitute the first instances of indirect selection. After all, complex quantitative traits in mature trees are forecasted by relying on measures of well-correlated traits earlier in the trees’ life cycle. However, with the advent of the genomic era, the speed and precision of indirect selection has been accentuated by screening the underlying genetic markers, enabling molecular breeders to trace and select for heritable additive variation rather than earlier trait segregation. At the dawn of the technology, predictions were built upon few associated markers linked with the desirable traits, an approach referred to as marker assisted selection (MAS). Meanwhile, increasing marker sets and sequenced genomes piled up, which made genome-wide predictions feasible as part of the modern genomic selection paradigm. When combined with other “omics” technologies such as transcriptomics, metabolomics, phenomics, and enviromics, the spectrum of inferences scales up, empowering more robust machine learning (ML)-based inferences across entire metabolic pathways, their phenotypic consequences, and the plant breeding triangle (genome + phenome + envirome). As if these progressions were not fascinating enough, tree breeders are starting to implement speed breeding, in which species’ life cycles are shorten even more by a combination of hormones, grafting, light treatments, growth chamber facilities, and gene editing of flowering genes. This way, precocious seedling and saplings can reach flowering maturity only few months into the greenhouse, allowing an expeditious first breeding generation. When coupled together, genomic selection and speed breeding promise quick and precise tree breeding cycles, easing population improvement at the pace imposed by climate change, environmental variation, business turnover, and extramural funding in tropical regions.