Genomic Selection a Powerful Tool for Maize Cultivar Adaptation to Climate Change

New genotyping platforms are coming on-stream that will rapidly change the nature of maize cultivar development. These methods have the potential to greatly facilitate the process of developing cultivars adapted to a changing climate, and will increase the importance of 'open-source' multinational testing networks.

It is estimated that, within maize breeding programmes, approximately between two and eight haplotypes per gene are present, and that many of these haplotypes recur across breeding programmes in different frequencies (Ching et al., 2002). A larger number of haplotypes per gene exists within the species, but this number is not infinite. Genotyping platforms are advancing rapidly in their ability to distinguish large numbers of haplotypes for dense genetic maps. A 60,000 single nucleotide polymorphism (SNP) genotyping array is currently available for maize that will permit genotyping of individual lines at high density for approximately US$200 per sample; the cost per sample for high-density genotyping is likely to fall by at least an order of magnitude over the next few years, while marker density increases by a similar amount.

The ability to obtain high-density geno-typic information inexpensively (i.e. for less than the cost of measuring yield in a single-location replicated field trial) means that it will become feasible to genotype at high density, and with the ability to discriminate among many alleles, all breeding lines enter ing replicated testing, permitting the ongoing estimation of haplotype allele effects from multi-location trials. Genotypic value of a line will be predicted, using methods first outlined by Meuwissen et al. (2001), from the combined value of marker haplo-type effects at intervals of 1 cM or less. This approach, referred to as genomic selection (GS), essentially treats the haplotype, rather than the line, as the selection unit. Simulation studies have shown that GS can accurately predict breeding values for quantitative traits, and the method is rapidly being applied in animal breeding (Hayes et al., 2009).

The shift from line to haplotype as selection unit made possible by GS means that a line does not need to be evaluated in an environment to predict its performance in that environment; rather, performance can be based on estimates of the effects of the haplotypes that comprise its genotype, estimated in other environments. For example if a specific haplotype allele effect is independently estimated in different backgrounds in several different breeding programmes serving a similar ME (e.g. low-rainfall subtropics), the value of that allele can be reasonably extrapolated to other, similar environments, in which breeding programmes are likely to be using similar germplasm. Haplotype effects of alleles in MEs into which a location or region is likely to shift due to climate change will also be available for prediction. Thus, using GS, it will be possible to: (i) predict the performance of lines under development for environments that are likely to occur with higher frequency in the future due to climate change; and (ii) rapidly identify, from testing networks in other environments, alleles that may be useful in coping with new conditions, or conditions that cannot be reliably tested for, in the TPE. Effective application of GS as a tool to cope with climate change will permit the linking of data from different multi-location testing networks based on genotypic information. Similar approaches are likely to be available in most crop species within the next 5 years.

0 0

Post a comment