2.04.02 - Breeding theory and progeny testing

UNIT NOTICEBOARD

2019-07-15

The IUFRO Seed Orchard conference 2019 - Seed Orchards and Climate Change

Nanjing, China; 14-16 October 2019. IUFRO Units involved: 2.09.01, 2.09.03, 2.04.02, 2.02.23, 2.02.07.

The goal of this conference is to provide the opportunity for scientists, students and managers of seed orchards and forest regeneration to exchange the most recent scientific advances related to forest tree seed orchards and their integration in the forestry practice. Areas that will be encompassed at the conference include, but are not limited to, the link between seed orchards and long-tem tree breeding; seed orchard design and management; forest pathology in relation to seed production, seed testing and storage; seed physiology and technology; forest economics; genetic resources and gene conservation; progeny testing; and interaction of seed orchards with related disciplines.

Details: http://seedorchard2019.njfu.edu.cn/

View all entries

Coordinator:

Jaroslav Klapste, New Zealand

Deputies:

Eduardo Pablo Cappa, Argentina

Gregory Dutkowski, Australia

About Unit

The Working Party is interested in research leading to the most effective methods for the structure and management of breeding programs. This includes the development of selection criteria and economic breeding objective traits, and the prediction of genetic values of genetic entries and gain from selection, as key aspects of tree breeding for both economically important traits and adaptability to future climatic change. As the forest tree breeding programmes have entered advanced generations of breeding cycles, and genotyping platforms have become well established and affordable, it is also concerned with the improvement of progeny testing methodology under the implementation of large datasets with complex pedigrees and genomic information.


State of Knowledge

The development of statistical approaches within the linear mixed model framework has allowed for the use of information about the relatedness between individual genotypes in populations based on documented pedigrees (Henderson 1986). This enabled the prediction of individual genotype breeding values, which in turn facilitated selection across generations to increase the response to selection (Ruotsalainen and Lindgren 1998). Forest tree breeding is different from animal breeding as genotypes can be cloned, can have many tested genotypes per family, they are tested in designed field trials and their age to final harvest and time to sexual maturity can be substantially more than the selection age. Additionally, indirect selection is usually practiced, as while early individual tree performance is measured, the objective trait for growth should be stand volume across multiple harvests, or often mixed genotype stands. Genotype x environment interaction (GxE) is one the most serious issues in the forest tree breeding, and the current rate of climate change puts even more emphasis on this issue. Therefore, an implementation of evaluation systems facilitating the exploitation of GxE is an urgent task for most breeding programmes. Cullis et al. (2014) proposed a factor analytic approach as a viable solution for large-scale multi-environment analysis in forest trees to understand and exploit GxE. Additionally, field experiments in forest tree breeding are generally large, with high microenvironmental heterogeneity. Approaches that account for spatial trends in natural and silvicultural variation are needed to account for this heterogeneity (Dutkowski et al. 2002). Since trees are long lived organisms, their survival and growth over time affects the growth of neighbouring trees and ultimately changes the estimates of their genetic values. Cappa and Cantet (2008) developed an approach to integrate competition effects into the linear mixed model framework. When both spatial heterogeneity and within tree competition are present, their considerations in the genetic evaluation allow for more precise separation of genetic and environmental effects and thus more accurate estimates of genetic parameters (Cappa et al. 2016).

Forest tree breeding programmes inevitably increase inbreeding as the programs progress, which in turn reduces genetic diversity and promotes the occurrence of inbreeding depression. Therefore, careful selection and mate allocation has to be carried out to balance gain against these detrimental consequences of increased inbreeding. Several approaches have been developed such as "Group merit selection" (Lindgren and Mullin 1997, Rosvall and Mullin 2003) or strategies based on quadratic optimization (Funda et al. 2009) maximizing genetic merit under constrained genetic diversity. This can be further expanded into targeted mating reflecting parent’s genetic worth such as stratified sublines (Ruotsalainen and Lindgren 2000) or through positive assortative mating with the unbalanced testing effort given the genetic worth of the cross (Lstiburek et al. 2005). Such approaches can generate additional additive genetic variance and thus the response to selection. However, the utilization of full potential generated from these advanced breeding approaches depends on the quality of selection indices properly combining multiple breeding objectives (Burdon and Klapste, 2019).

Current progress in genotyping platforms has allowed for development of genomic resources based on genotyping-by-sequencing (Elshire et al. 2011) or exome capture (Neves et al. 2013) also in forest trees. Pedigree reconstruction using genetic markers has been broadly implemented in forest tree breeding populations as well as in forest plantations to recover hidden relatedness (effectively converting open pollinated to control pollinated families) or to correct pedigree errors and usually results in improved management of genetic diversity and the accuracy of genetic parameters and breeding values (Doerksen and Herbinger 2010, Hansen and McKinney 2010, El-Kassaby et al. 2011, Bouffier et al. 2019, Lstiburek et al. 2020). This might allow for reduction in testing effort as less control pollinated trials are needed. Additionally, genetic marker information can be integrated with pedigree records in single-step genomic evaluations by augmenting the relationship matrix used in the evaluation. This gives an increase in accuracy in the predicted breeding values for genotypes with phenotype information (Ratcliffe et al. 2017, Cappa et al. 2017, Cappa et al. 2018) and for genotypes with no phenotypic information. This allows early selection in new generation breeding material and the faster delivery of genetically improved material to commercial plantations (Grattapaglia et al. 2018, Cappa et al. 2019).

For further publications, please visit https://www.iufro.org/science/divisions/division-2/20000/20400/20402/publications/.