Sugarcane cultivars are interspecific hybrids with an aneuploid, highly heterozygous polyploid genome. The complexity of the sugarcane genome is the main obstacle to the use of marker-assisted selection in sugarcane breeding. Given the promising results of recent studies of plant genomic selection, we explored the feasibility of genomic selection in this complex polyploid crop. Genetic values were predicted in two independent panels, each composed of 167 accessions representing sugarcane genetic diversity worldwide. Accessions were genotyped with 1,499 DArT markers. One panel was phenotyped in Reunion Island and the other in Guadeloupe. Ten traits concerning sugar and bagasse contents, digestibility and composition of the bagasse, plant morphology, and disease resistance were used. We used four statistical predictive models: bayesian LASSO, ridge regression, reproducing kernel Hilbert space, and partial least square regression. The accuracy of the predictions was assessed through the correlation between observed and predicted genetic values by cross validation within each panel and between the two panels. We observed equivalent accuracy among the four predictive models for a given trait, and marked differences were observed among traits. Depending on the trait concerned, within-panel cross validation yielded median correlations ranging from 0.29 to 0.62 in the Reunion Island panel and from 0.11 to 0.5 in the Guadeloupe panel. Cross validation between panels yielded correlations ranging from 0.13 for smut resistance to 0.55 for brix. This level of correlations is promising for future implementations. Our results provide the first validation of genomic selection in sugarcane.

The sugarcane variety development program on Réunion Island is dedicated to an industry that encompasses numerous different agroclimatic production zones. The objective of this study is to characterize in detail the final selection stage of this program, consisting of multienvironment trials (MET) at seven representative locations, considering the genotypic response in terms of tonnes of cane per hectare (TCH), estimable recoverable sugar (ERS), fiber content (FIB), and an economic index (EI). Data from four recent variety series tested between one and three crop years were used. Each trait revealed a significant genotype 9 location (GL) interaction, always higher than the genotype 9 crop year (GC) interaction, indicating that testing genotypes across locations is more important than testing for ratooning ability. Broad-sense heritability (H) at MET level was relatively high for FIB compared with EI, while ERS and TCH exhibited intermediate H values. Genotype main effect plus genotype 9 environment (GGE) biplot analysis applied to a balanced set of genotypes tested over two crop years across all environments permitted reliable visualization at a glance of (i) the level of proximity between genotypes or environments, and (ii) the performance of varieties in each environment and their stability across all of them. No redundancy between any pair of environments was found for the most important selection trait (EI). These results confirm the relevance of a selection strategy firstly oriented toward selecting sugarcane genotypes for local adaptations with the objective of enhancing the mean productivity of the whole cane industry.

Sugarcane variety development programmes are costly and lengthy. It appears important to periodically assess their ability to select competitive genotypes for yield components and provide genetic gains. This article reports a 10-year retrospective analysis of successive variety trials conducted in four regional breeding programmes on Réunion Island in the most advanced regional selection stage. The four variety programmes were dedicated to the humid coastal zone (LM), the per-humid coastal zone (SB), the irrigated dry coastal zone (ES) and the dry high lands (VB) of the local industry. Using mixed linear models, the objective of the study was to (1) assess genetic variabilities available for yield components in this advanced selection stage; and (2) estimate trends of genetic gains achieved over the last decade across the four programmes. Yield components were: cane yield (CY), estimable recoverable sugar (ERS), fibre content (FIB) and economic index (EI). Broad-sense heritabilities were high (0.70-0.91) for all traits in each programme. Mean genetic coefficients of variation were about twice as high for CY (14.1%) and EI (15.5%) compared to ERS (5.7%) and FIB (6.6%). A higher probability of identifying superior varieties was found for CY and EI in two of the four programmes characterized either by thermal and hydric stresses (VB) or by an edaphic stress (SB). Simple linear regression of variety performances versus years of selection revealed trends in genetic gains for EI ranging between 0.53 and 1.81% increase per year that were highly significant (P \ 0.001) in the two programmes (LM and VB).

All over the world, sugarcane breeding programs are developing new, high-yielding cultivars that are resistant to major diseases to improve the profitability and sustainability of the sugar-energy industries they serve. In Reunion Island, sugarcane genetic improvement efforts began in 1929. Many challenges had to be overcome. Continuous breeding efforts have been made to develop varietal resistances to control some major diseases and are still going on today. Given the extreme agroclimatic diversity that characterizes the different production areas of the industry, it was necessary to gradually develop a large network of seven decentralized breeding programs to support genetic progress throughout the whole industry. This article provides an overview of the sugarcane breeding program of Reunion. It describes historical achievements and gives detailed information about germplasm development, variety exchanges, breeding program and selection scheme and procedures. A review is also made on applied genetics research activities supporting variety improvement. Further progress depends on the optimized functioning of the current breeding program, which has never been so largely extended in terms of target environments. The article discusses prospects of genomics breeding applications in the complex genetic context of sugarcane, which will require large multidisciplinary collaborations.

No abstract