A demonstration of the enviromics approach to integrating environmental ‘big data’ problems DOI Open Access
Andrew F. Bowerman

New Phytologist, Год журнала: 2024, Номер unknown

Опубликована: Авг. 30, 2024

'The ability to accurately predict crop performance in untested environments holds substantial promise for addressing the challenges of global food security…' Resende et al.'s research is particularly significant given extensive geographical range and environmental variability within which maize cultivated. Indeed, measures ambitious: 183 field trials conducted across four Brazilian states, involving 79 phenotyped hybrids their 85 nonphenotyped parents. Data collection was carried out from 2017 2021, encompassing various covariates sourced weather, soil, sensors, satellites, adding up over 1300 envirotypic covariates. By focusing on precise characterization, study aimed optimize breeding high-yielding, stable hybrids. The concepts envirotypes enviromics are relatively new, certainly as applied plant efforts (Costa-Neto & Fritsche-Neto, 2021; al., 2021). Enviromics a that integrates data with genomics better understand interactions between an organism's genetic makeup its environment. This interdisciplinary approach leverages variety mentioned above how these factors collectively influence phenotypic traits. most recent (2024b) extends proposed methodology (Resende 2024a) use Geographic Information Systems (GIS) platform purpose high-density envirotyping. GIS environment this meticulously designed include geoprocessing polygon all experimental points, 50 km buffer zone ensure comprehensive coverage. setup resulted prediction grid comprising 14 966 bins covering states São Paulo, Paraná, Santa Catarina Rio Grande de Sul. These represent considerable variation, Paulo's diverse tropical temperate climate, Paraná's subtropical Catarina's mix coastal highland climates strong industrial tourism sectors, do Sul's climate. A key contribution development Engineered Enviromic Markers (EEM), provide novel understanding predicting hybrid performance. aggregating predictors into Random Forests using hierarchical clustering, researchers created robust model capable handling complexities G × E interactions. also introduced Reaction (REEM) model, ensemble modelling technique combines predictions multiple models enhance overall predictive accuracy. correlations derived allowed define Breeding Zones studied, then yield genotype onto map. Yield stability could be predicted by strength relationship EEMs, low relationships indicating greater stability. Forest methodologies employed EEMs interesting they largely agnostic (i.e. nature different sets included irrelevant) system allows any new type future. types radar, thermal or LiDAR sensors (Newman Furbank, 2024a), or, indeed technologies not yet employed. Already has used MODIS, WorldClim, SoilGrid NASA Power build integration precision represents advancement agricultural science. providing detailed characterizations leveraging machine learning techniques, al. offer pathway more efficient effective strategies. Current approaches genotype–environment interactions, especially agronomic conditions, require broad environments. Trials can prohibitively expensive often aren't feasible, but collate here demonstrates value small farm-based records modelling. security, face climate change (Fig. 1). What findings parental lines should generate maximize area; genotypes have been trialled area question. results tested published. Detailed enable breeders make informed decisions, ultimately leading better-performing crops tailored specific conditions. In future, it will compelling see developed real-time data, through cropping cycle updates end season yields. Expanding species engineered enviromic markers/environmental preferences important evaluate wider value, differences (Swarup Khoury 2022). future deployment methodologies, amply demonstrated (2024b), allow far estimation due variance strengthen productivity.

Язык: Английский

Leveraging Automated Machine Learning for Environmental Data‐Driven Genetic Analysis and Genomic Prediction in Maize Hybrids DOI Creative Commons
Kunhui He,

Tingxi Yu,

Shang Gao

и другие.

Advanced Science, Год журнала: 2025, Номер unknown

Опубликована: Март 6, 2025

Genotype, environment, and genotype-by-environment (G×E) interactions play a critical role in shaping crop phenotypes. Here, large-scale, multi-environment hybrid maize dataset is used to construct validate an automated machine learning framework that integrates environmental genomic data for improved accuracy efficiency genetic analyses predictions. Dimensionality-reduced parameters (RD_EPs) aligned with developmental stages are applied establish linear relationships between RD_EPs traits assess the influence of environment on phenotype. Genome-wide association study identifies 539 phenotypic plasticity trait-associated markers (PP-TAMs), 223 stability TAMs (Main-TAMs), 92 G×E-TAMs, revealing distinct bases PP G×E interactions. Training prediction models both increase by 14.02% 28.42% over genome-wide marker approaches. These results demonstrate potential utilizing improving analysis selection, offering scalable approach developing climate-adaptive varieties.

Язык: Английский

Процитировано

1
Accuracy of Multi‐Environmental Trials in Predicting New Environments Using Different Approaches Based on Environmental Covariates: A Case in Barley (Hordeum vulgare L.) Breeding DOI Creative Commons
Diriba Tadese, Hans‐Peter Piepho, Girma F. Dinsa

и другие.

Plant Breeding, Год журнала: 2025, Номер unknown

Опубликована: Март 2, 2025

ABSTRACT One of the current innovations in predicting genotype performances a target population environments is integrating environmental covariates (ECs) into multi‐environment trial (MET) data analysis. In this study, MET set barley ( Hordeum vulgare L.) breeding program years 2016 and 2017 was used. We evaluated compared different approaches using ECs new environments. The comparison done mean squared error predicted differences (MSEPD) under linear mixed models. MSEPD computed for cross‐validation mechanism that drops out one environment at time. Our results show models with resulted smaller model without ECs. Among approaches, reduced rank regression approach component smallest followed by fitting both first second synthetic extended Finlay–Wilkinson regression. Overall, there potential gain predictive accuracy plant programs.

Язык: Английский

Процитировано

0
GIS‐based G × E modeling of maize hybrids through enviromic markers engineering DOI Creative Commons
Rafael Tassinari Resende, Alencar Xavier, Pedro Italo T. Silva

и другие.

New Phytologist, Год журнала: 2024, Номер unknown

Опубликована: Июль 16, 2024

Through enviromics, precision breeding leverages innovative geotechnologies to customize crop varieties specific environments, potentially improving both yield and genetic selection gains. In Brazil's four southernmost states, data from 183 distinct geographic field trials (also accounting for 2017-2021) covered information on 164 genotypes: 79 phenotyped maize hybrid genotypes grain their 85 nonphenotyped parents. Additionally, 1342 envirotypic covariates weather, soil, sensor-based, satellite sources were collected engineer 10 K synthetic enviromic markers via machine learning. Soil, radiation light, surface temperature variations remarkably affect differential genotype yield, hinting at ecophysiological adjustments including evapotranspiration photosynthesis. The ensemble-based random regression model showcases superior predictive performance efficiency compared the baseline kernel models, matching best coordinates. Clustering analysis has identified regions that minimize genotype-environment (G × E) interactions. These findings underscore potential of enviromics in crafting parental combinations breed new, higher-yielding crops. adequate use can enhance by providing important inputs about environmental factors average performance. Generating associated with enable a better hybrids environments.

Язык: Английский

Процитировано

2
Accuracy of prediction from multi-environment trials for new locations using pedigree information and environmental covariates: the case of sorghum (Sorghum bicolor (L.) Moench) breeding DOI Creative Commons
Diriba Tadese, Hans‐Peter Piepho, Jens Hartung

и другие.

Theoretical and Applied Genetics, Год журнала: 2024, Номер 137(8)

Опубликована: Июль 10, 2024

Abstract Key messages We investigate a method of extracting and fitting synthetic environmental covariates pedigree information in multilocation trial data analysis to predict genotype performances untested locations. Plant breeding trials are usually conducted across multiple testing locations the targeted population environments. The predictive accuracy can be increased by use adequate statistical models. compared linear mixed models with without (SCs) under identity, diagonal factor-analytic variance-covariance structures genotype-by-location interactions. A comparison was made evaluate different predicting using mean squared error predicted differences (MSEPD) Spearman rank correlation between adjusted means. multi-environmental (MET) dataset evaluated for yield performance dry lowland sorghum ( Sorghum bicolor L. ) Moench program Ethiopia used. For validating our models, we followed leave-one-location-out cross-validation strategy. total 65 (ECs) obtained from test were considered. SCs extracted ECs multivariate partial least squares subsequently fitted model. Then, model extended accounting information. According MSEPD, SC improve three others SC. also higher When fitted, 0.58 factor analytic, 0.51 0.46 identity structures. Our approach indicates improvement context interactions Ethiopia.

Язык: Английский

Процитировано

1
Factor analytic selection tools and environmental feature-integration enable holistic decision-making in Eucalyptus breeding DOI Creative Commons
Saulo Fabrício da Silva Chaves, Michelle B. Damacena,

Kaio Olimpio G. Dias

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Авг. 8, 2024

Understanding the genotype-by-environment interaction (GEI) and considering it in selection process is a sine qua non condition for expansion of Brazilian eucalyptus silviculture. This study's objective to select high-performance stable clones based on novel index that considers Factor Analytic Selection Tools (FAST) clone's reliability. The investigation explores nuances interplay GEI extends its insights by scrutinizing relationship between latent factors real environmental features. analysis, conducted across seven trials five states involving 78 clones, employs FAST. clonal was performed using an extended FAST weighted Further about emerge from integration factor loadings with 25 features through principal component analysis. Ten distinguished high performance, stability, reliability, have been selected target population environments. most closely associated loadings, encompassing air temperature, radiation, soil characteristics, as pivotal drivers within this dataset. study contributes breeders, equipping them enhance decision-making harnessing holistic understanding-from genotypes under evaluation diverse environments anticipated commercial plantations.

Язык: Английский

Процитировано

1
Harnessing enviromics to predict climate‐impacted high‐profile traits to assist informed decisions in agriculture DOI Creative Commons

Bosen Zhang,

Amber L. Hauvermale, Zhiwu Zhang

и другие.

Food and Energy Security, Год журнала: 2024, Номер 13(3)

Опубликована: Май 1, 2024

Abstract Modern agriculture is a complex system that demands real‐time and large‐scale quantification of trait values for evidence‐based decisions. However, high‐profile traits determining market often lack high‐throughput phenotyping technologies to achieve this objective; therefore, risks undermining crop through arbitrary decisions are high. Because environmental conditions major contributors performance fluctuation, with the contemporary informatics infrastructures, we proposed enviromic prediction as potential strategy assess informed We demonstrated concept wheat falling number (FN), critical end‐use quality significantly impacts but measured using low‐throughput technology. Using 8 years FN records from elite variety testing trials, developed predictive model capturing general trend based on biologically meaningful conditions. An explicit index was highly correlated ( r = 0.646) observed trials identified. independent validation experiment verified biological relevance index. achieved accurate on‐target predictions in new growing seasons. Two applications designed production fields illustrated how such models could assist decision along food supply chain. envision would have vital role sustaining security amidst rapidly changing climate. As conducting standard component modern agricultural industry, leveraging historical trial data widely applicable other various crops.

Язык: Английский

Процитировано

0
Digital Twins: A Next‐Gen Solution for Agricultural Sustainability DOI

Abhishek Panchadi,

Bipin Bastakoti,

Prathiksha Raghava

и другие.

CSA News, Год журнала: 2024, Номер 69(9), С. 32 - 36

Опубликована: Авг. 14, 2024

Язык: Английский

Процитировано

0
A demonstration of the enviromics approach to integrating environmental ‘big data’ problems DOI Open Access
Andrew F. Bowerman

New Phytologist, Год журнала: 2024, Номер unknown

Опубликована: Авг. 30, 2024

'The ability to accurately predict crop performance in untested environments holds substantial promise for addressing the challenges of global food security…' Resende et al.'s research is particularly significant given extensive geographical range and environmental variability within which maize cultivated. Indeed, measures ambitious: 183 field trials conducted across four Brazilian states, involving 79 phenotyped hybrids their 85 nonphenotyped parents. Data collection was carried out from 2017 2021, encompassing various covariates sourced weather, soil, sensors, satellites, adding up over 1300 envirotypic covariates. By focusing on precise characterization, study aimed optimize breeding high-yielding, stable hybrids. The concepts envirotypes enviromics are relatively new, certainly as applied plant efforts (Costa-Neto & Fritsche-Neto, 2021; al., 2021). Enviromics a that integrates data with genomics better understand interactions between an organism's genetic makeup its environment. This interdisciplinary approach leverages variety mentioned above how these factors collectively influence phenotypic traits. most recent (2024b) extends proposed methodology (Resende 2024a) use Geographic Information Systems (GIS) platform purpose high-density envirotyping. GIS environment this meticulously designed include geoprocessing polygon all experimental points, 50 km buffer zone ensure comprehensive coverage. setup resulted prediction grid comprising 14 966 bins covering states São Paulo, Paraná, Santa Catarina Rio Grande de Sul. These represent considerable variation, Paulo's diverse tropical temperate climate, Paraná's subtropical Catarina's mix coastal highland climates strong industrial tourism sectors, do Sul's climate. A key contribution development Engineered Enviromic Markers (EEM), provide novel understanding predicting hybrid performance. aggregating predictors into Random Forests using hierarchical clustering, researchers created robust model capable handling complexities G × E interactions. also introduced Reaction (REEM) model, ensemble modelling technique combines predictions multiple models enhance overall predictive accuracy. correlations derived allowed define Breeding Zones studied, then yield genotype onto map. Yield stability could be predicted by strength relationship EEMs, low relationships indicating greater stability. Forest methodologies employed EEMs interesting they largely agnostic (i.e. nature different sets included irrelevant) system allows any new type future. types radar, thermal or LiDAR sensors (Newman Furbank, 2024a), or, indeed technologies not yet employed. Already has used MODIS, WorldClim, SoilGrid NASA Power build integration precision represents advancement agricultural science. providing detailed characterizations leveraging machine learning techniques, al. offer pathway more efficient effective strategies. Current approaches genotype–environment interactions, especially agronomic conditions, require broad environments. Trials can prohibitively expensive often aren't feasible, but collate here demonstrates value small farm-based records modelling. security, face climate change (Fig. 1). What findings parental lines should generate maximize area; genotypes have been trialled area question. results tested published. Detailed enable breeders make informed decisions, ultimately leading better-performing crops tailored specific conditions. In future, it will compelling see developed real-time data, through cropping cycle updates end season yields. Expanding species engineered enviromic markers/environmental preferences important evaluate wider value, differences (Swarup Khoury 2022). future deployment methodologies, amply demonstrated (2024b), allow far estimation due variance strengthen productivity.

Язык: Английский

Процитировано

0