GenoDrawing: An Autoencoder Framework for Image Prediction from SNP Markers

Integrative multi-environmental genomic prediction in apple

Genomic prediction for multiple environments can aid the selection of genotypes suited to specific soil and climate conditions. Methodological advances allow effective integration of phenotypic, genomic (additive, nonadditive), and large-scale environmental (enviromic) data into multi-environmental genomic prediction models. These models can also account for genotype-by-environment interaction, utilize alternative relationship matrices (kernels), or substitute statistical approaches with deep learning. However, the application of multi-environmental genomic prediction in apple remained limited, likely due to the challenge of building multi-environmental datasets and structurally complex models. Here, we applied efficient statistical and deep learning models for multi-environmental genomic prediction of eleven apple traits with contrasting genetic architectures by integrating genomic- and enviromic-based model components. Incorporating genotype-by-environment interaction effects into statistical models improved predictive ability by up to 0.08 for nine traits compared to the benchmark model. This outcome, based on Gaussian and Deep kernels, shows these alternatives can effectively substitute the standard genomic best linear unbiased predictor (G-BLUP). Including nonadditive and enviromic-based effects resulted in a predictive ability very similar to the benchmark model. The deep learning approach achieved the highest predictive ability for three traits with oligogenic genetic architectures, outperforming the benchmark by up to 0.10. Our results demonstrate that the tested statistical models capture genotype-by-environment interactions particularly well, and the deep learning models efficiently integrate data from diverse sources. This study will foster the adoption of multi-environmental genomic prediction to select apple cultivars adapted to diverse environmental conditions, providing an opportunity to address climate change impacts.

Genomic analysis of fruit size and shape traits in apple: unveiling candidate genes through GWAS analysis

Genomic tools facilitate the efficient selection of improved genetic materials within a breeding program. Here, we focus on two apple fruit quality traits: shape and size. We utilized data from 11 fruit morphology parameters gathered across three years of harvest from 355 genotypes of the apple REFPOP collection, which serves as a representative sample of the genetic variability present in European-cultivated apples. The data were then employed for genome-wide association studies (GWAS) using the FarmCPU and the BLINK models. The analysis identified 59 SNPs associated with fruit size and shape traits (35 with FarmCPU and 45 with BLINK) responsible for 71 QTNs. These QTNs were distributed across all chromosomes except for chromosomes 10 and 15. Thirty-four QTNs, identified by 27 SNPs, were related for size traits, and 37 QTNs, identified by 26 SNPs, were related to shape attributes. The definition of the haploblocks containing the most relevant SNPs served to propose candidate genes, among them the genes of the ovate family protein MdOFP17 and MdOFP4 that were in a 9.7kb haploblock on Chromosome 11. RNA-seq data revealed low or null expression of these genes in the oblong cultivar "Skovfoged" and higher expression in the flat "Grand'mere." The Gene Ontology enrichment analysis support a role of OFPs and hormones in shape regulation. In conclusion, this comprehensive GWAS analysis of the apple REFPOP collection has revealed promising genetic markers and candidate genes associated with apple fruit shape and size attributes, providing valuable insights that could enhance the efficiency of future breeding programs.