Partial least squares enhance multi-trait genomic prediction of potato cultivars in new environments

Ortíz, Rodomiro; Reslow, Fredrik; Montesinos‐López, Abelardo; Huicho, José; Pérez‐Rodríguez, Paulino; Montesinos‐López, Osval A.; Crossa, J.

doi:10.1038/s41598-023-37169-y

Cited by 7 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notably, we observed a substantial difference between genomic and pedigree heritability estimates for dry matter content, a trait known to be highly heritable (Ortiz et al, 2023). This indicates that some genetic variance is not captured in the G-matrix-based model employed in GS.…”

Section: Dry Matter Content Heritability Estimate Discrepancymentioning

confidence: 83%

“…In this study, we used a single-trait standard additive GBLUP model for evaluating the impact of marker type and marker density on prediction performance since using a simple model on the individual traits facilitated the interpretation of these effects. However, in breeding programs, clones are selected on multiple traits concurrently, and in such a case, multi-trait models are more suitable to use for optimizing genomic prediction models to support selection (Ortiz et al, 2023).…”

Section: Consequences For Breedingmentioning

confidence: 99%

“…GS is particularly useful for traits with complex, multigenic inheritance patterns, such as chipping quality and dry matter content, where optimization has been largely ineffective in traditional breeding by phenotype-directed sexual crosses (Wilson et al, 2021). Several studies have already evaluated the performance of different statistical methods and machine learning approaches for the reliable prediction of genomic estimated breeding values for a host of agronomic performance traits, training on a vast collection of elite breeding materials (Slater et al, 2016;Sverrisdottir et al, 2017Sverrisdottir et al, , 2018Stich and Van Inghelandt, 2018;Byrne et al, 2020;Selga et al, 2021;Wilson et al, 2021;Ortiz et al, 2022Ortiz et al, , 2023Pandey et al, 2023). However, the development of strategies for enhancing the power of GS is still highly relevant for potato breeding.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The effect of marker types and density on genomic prediction and GWAS of key performance traits in tetraploid potato

Aalborg,

Sverrisdóttir,

Kristensen

et al. 2024

Front. Plant Sci.

View full text Add to dashboard Cite

Genomic prediction and genome-wide association studies are becoming widely employed in potato key performance trait QTL identifications and to support potato breeding using genomic selection. Elite cultivars are tetraploid and highly heterozygous but also share many common ancestors and generation-spanning inbreeding events, resulting from the clonal propagation of potatoes through seed potatoes. Consequentially, many SNP markers are not in a 1:1 relationship with a single allele variant but shared over several alleles that might exert varying effects on a given trait. The impact of such redundant “diluted” predictors on the statistical models underpinning genome-wide association studies (GWAS) and genomic prediction has scarcely been evaluated despite the potential impact on model accuracy and performance. We evaluated the impact of marker location, marker type, and marker density on the genomic prediction and GWAS of five key performance traits in tetraploid potato (chipping quality, dry matter content, length/width ratio, senescence, and yield). A 762-offspring panel of a diallel cross of 18 elite cultivars was genotyped by sequencing, and markers were annotated according to a reference genome. Genomic prediction models (GBLUP) were trained on four marker subsets [non-synonymous (29,553 SNPs), synonymous (31,229), non-coding (32,388), and a combination], and robustness to marker reduction was investigated. Single-marker regression GWAS was performed for each trait and marker subset. The best cross-validated prediction correlation coefficients of 0.54, 0.75, 0.49, 0.35, and 0.28 were obtained for chipping quality, dry matter content, length/width ratio, senescence, and yield, respectively. The trait prediction abilities were similar across all marker types, with only non-synonymous variants improving yield predictive ability by 16%. Marker reduction response did not depend on marker type but rather on trait. Traits with high predictive abilities, e.g., dry matter content, reached a plateau using fewer markers than traits with intermediate-low correlations, such as yield. The predictions were unbiased across all traits, marker types, and all marker densities >100 SNPs. Our results suggest that using non-synonymous variants does not enhance the performance of genomic prediction of most traits. The major known QTLs were identified by GWAS and were reproducible across exonic and whole-genome variant sets for dry matter content, length/width ratio, and senescence. In contrast, minor QTL detection was marker type dependent.

show abstract

Section: Dry Matter Content Heritability Estimate Discrepancymentioning

confidence: 83%

Section: Consequences For Breedingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The effect of marker types and density on genomic prediction and GWAS of key performance traits in tetraploid potato

Aalborg,

Sverrisdóttir,

Kristensen

et al. 2024

Front. Plant Sci.

View full text Add to dashboard Cite

show abstract

“…With the advancement of computational technology and the democratization of "enviromics" resources, PLS became a suitable method for exploring big data to perform spatial predictions of experimental genotypes in new environments (Monteverde et al, 2019;Rincent et al, 2019;Guo et al, 2021;Costa-Neto et al, 2022). In fact, PLS has emerged as a relevant alternative for prediction purposes, even when breeders do not specifically incorporate environmental data into the model (Montesinos-López et al, 2022b,a;Ortiz et al, 2023).…”

Section: Predicting Using Partial Least Squares Regressionmentioning

confidence: 99%

GIS-FA: an approach to integrating thematic maps, factor-analytic, and envirotyping for cultivar targeting

Araújo,

Chaves,

Dias

et al. 2024

Theor Appl Genet

View full text Add to dashboard Cite

Key message: We propose an enviromics prediction model for cultivar recommendation based on thematic maps for decision-makers. Parsimonious methods that capture genotype-by-environment interaction (GEI) in multi-environment trials (MET) are important in breeding programs. Understanding the causes and factors of GEI allows the utilization of genotype adaptations in the target population of environments through environmental 1 features and Factor-Analytic (FA) models. Here, we present a novel predictive breeding approach called GIS-FA that integrates geographic information systems (GIS) techniques, FA models, Partial Least Squares (PLS) regression, and Enviromics to predict phenotypic performance in untested environments. The GIS-FA approach allows: (i) predict the phenotypic performance of tested genotypes in untested environments; (ii) select the best-ranking genotypes based on their overall performance and stability using the FA selection tools; (iii) draw thematic maps showing overall or pairwise performance and stability for decision-making. We exemplify the usage of GIS-FA approach using two datasets of rice [Oryza sativa (L.)] and soybean [Glycine max (L.) Merr.] in MET spread over tropical areas. In summary, our novel predictive method allows the identification of new breeding scenarios by pinpointing groups of environments where genotypes have superior predicted performance and facilitates/optimizes the cultivar recommendation by utilizing thematic maps.

show abstract

“…With the advancement of computational technology and the democratization of "enviromics" resources, PLS has emerged as a suitable method for exploring big data and performing spatial predictions of experimental genotypes in new environments (MONTEVERDE et al, 2019;RINCENT et al, 2019;GUO et al, 2021;COSTA-NETO et al, 2022). In fact, PLS has emerged as a relevant alternative for prediction purposes, even when breeders do not specifically incorporate environmental data into the model (MONTESINOS-LOPEZ et al, 2022b;MONTESINOS-LOPEZ et al, 2022a;ORTIZ et al, 2023).…”

Section: Predicting Using Partial Least Squares Regressionmentioning

confidence: 99%

Integrating statistical genetics, geographical information systems and envirotyping: a novel approach for predictive breeding and decision-making

Araújo

View full text Add to dashboard Cite

The crossover genotype-by-environments (G × E) interaction is responsible for the variation in genotype performance across diﬀerent environments. Disregarding the eﬀect of this inter- action means neglecting the speciﬁc adaptations of genotypes in the target population of environments. Environmental characterization enables an understanding of the speciﬁcities and similarities among diﬀerent environments. In this context, enviromics has emerged as a new area that integrates information from data analysis, quantitative genetics, geographic information systems (GIS), and principles of ecophysiology. Incorporating environmental features into Statistical Genetics models contributes to enhancing the predictive capability of these models and a better understanding of the cultivation environment. Thus, this study aimed to propose a new predictive breeding method (GIS-FA) that integrates GIS information, factor analytic (FA) models, partial least squares regression (PLS), and enviromics to predict purelines of rice and soybean in untested environments. Two databases were used: one for rice, with 80 purelines cultivated in the years 2009/10 and 2010/11 in 21 environments in eight Brazilian states; and the second composed of 195 soybean purelines evaluated in 49 environments (in years 2019/20, 2020/21, and 2021/22) in the state of Mato Grosso do Sul. The term “environment” refers to the location-year combination. For both datasets, FA models were adjusted, with FA4 being selected based on the average semivariance ratio. A total of 32 environmental features (EF) were collected, including three geographical, 16 climatic, and 13 soil-related features. To make predictions, 50 points were randomly chosen within each municipality in the evaluated states, and, for each point, EFs data were obtained from a historical series (2000-2021). Leveraging the FA model outcomes, we used the PLS method to predict the overall performance and stability of both crops in untested environments. Cross-validation was performed using the leave-one-out method, and subsequently, the GIS-FA method was compared with the GGE-GIS approach, which uses directly the within-environment eBLUPs to perform the prediction. After the spatial prediction, performance and stability parameters were represented in thematic maps. For predicting eBLUEs, GIS-FA was 10% and 1% superior to GIS-GGE in the rice and soybean datasets, respectively. For predicting eBLUPs, GIS-FA was 9% and 5% more eﬀective than GIS-GGE. Three types of maps were created: (i) zones of genotype adaptation; (ii) pairwise comparison between pureline vs. check and pureline vs. pureline; (iii) which-won-where. The GIS-FA approach proved to be eﬃcient in predicting genotypes for untested environments, allowing the evaluation of the G × E interaction throughout the experimental network. Keywords: Environmental features. Factor analytic. Predictive models. Partial Least Squares.

show abstract

Partial least squares enhance multi-trait genomic prediction of potato cultivars in new environments

Cited by 7 publications

References 46 publications

The effect of marker types and density on genomic prediction and GWAS of key performance traits in tetraploid potato

The effect of marker types and density on genomic prediction and GWAS of key performance traits in tetraploid potato

GIS-FA: an approach to integrating thematic maps, factor-analytic, and envirotyping for cultivar targeting

Integrating statistical genetics, geographical information systems and envirotyping: a novel approach for predictive breeding and decision-making

Contact Info

Product

Resources

About