Adam Norman
Research Subjects
Adam Norman has a total of 41 co-authors. Here are the top co-authors by number of shared publications:
Haydn Kuchel
Centre for Biometrics and Data Science for Sustainable Primary Industries, School of Mathematics and Applied Statistics, National Institute for Applied Statistics Research Australia, University of Wollongong, Wollongong, NSW, Australia
3 shared publicationsJames Edwards
School of Agriculture, Food and Wine, The University of Adelaide, Adelaide, Australia
3 shared publications- Julian Taylor
Waite Research Institute, School of Agriculture, Food and Wine, Adelaide University , Waite Campus, Adelaide
3 shared publications - Alison B. Smith
University of Wollongong
1 shared publication - Alison Barbara Smith
University of Wollongong
1 shared publication - Andrew Gock
CSIRO Environment
1 shared publication
Latest Publications
- Open Access
OzWheat: a genome-to-phenome platform to resolve complex traits for wheat pre-breeding and research
For over a century, Australian wheat breeders have successfully adapted wheat to a broad range of climatic conditions and crop management practices. The OzWheat genome-to-phenome (G2P) platform was established to capture this breeding history and explore traits, genes, and their interactions with the environment to enable ongoing research and deliver targets for wheat improvement. A panel of 285 cultivars and landraces were chosen through knowledge of breeding pedigrees to represent both global diversity and the historic flow of genetic variation over more than 100 years of selective breeding in Australia. Genetic characterisation of the panel included identification of genome-wide sequence variants and gene expression profiling across environments. Important traits for adaptation (flowering time and plant height) were assayed in controlled environments and at multiple field sites and years, with genome-wide association analyses (GWAS) using linear mixed models detecting both known and novel loci. Here, we report establishment of the OzWheat G2P platform as a powerful tool to integrate wheat genomes and phenomes and demonstrate its use to identify candidate genes and understand gene by environment interactions. This provides the wheat research and breeding community a new resource to support future cultivar development.
Open AccessOzWheat: a genome-to-phenome platform to resolve complex traits for wheat pre-breeding and research
For over a century, Australian wheat breeders have successfully adapted wheat to a broad range of climatic conditions and crop management practices. The OzWheat genome-to-phenome (G2P) platform was established to capture this breeding history and explore traits, genes, and their interactions with the environment to enable ongoing research and deliver targets for wheat improvement. A panel of 285 cultivars and landraces were chosen through knowledge of breeding pedigrees to represent both global diversity and the historic flow of genetic variation over more than 100 years of selective breeding in Australia. Genetic characterisation of the panel included identification of genome-wide sequence variants and gene expression profiling across environments. Important traits for adaptation (flowering time and plant height) were assayed in controlled environments and at multiple field sites and years, with genome-wide association analyses (GWAS) using linear mixed models detecting both known and novel loci. Here, we report establishment of the OzWheat G2P platform as a powerful tool to integrate wheat genomes and phenomes and demonstrate its use to identify candidate genes and understand gene by environment interactions. This provides the wheat research and breeding community a new resource to support future cultivar development.
- Open Access
- 2 September 2023
- Published by Springer Nature in Plant Methods
- Vol. 19 (1) , 1-14
Improved genomic prediction using machine learning with Variational Bayesian sparsity
Background: Genomic prediction has become a powerful modelling tool for assessing line performance in plant and livestock breeding programmes. Among the genomic prediction modelling approaches, linear based models have proven to provide accurate predictions even when the number of genetic markers exceeds the number of data samples. However, breeding programmes are now compiling data from large numbers of lines and test environments for analyses, rendering these approaches computationally prohibitive. Machine learning (ML) now offers a solution to this problem through the construction of fully connected deep learning architectures and high parallelisation of the predictive task. However, the fully connected nature of these architectures immediately generates an over-parameterisation of the network that needs addressing for efficient and accurate predictions. Results: In this research we explore the use of an ML architecture governed by variational Bayesian sparsity in its initial layers that we have called VBS-ML. The use of VBS-ML provides a mechanism for feature selection of important markers linked to the trait, immediately reducing the network over-parameterisation. Selected markers then propagate to the remaining fully connected feed-forward components of the ML network to form the final genomic prediction. We illustrated the approach with four large Australian wheat breeding data sets that range from 2665 lines to 10375 lines genotyped across a large set of markers. For all data sets, the use of the VBS-ML architecture improved genomic prediction accuracy over legacy linear based modelling approaches. Conclusions: An ML architecture governed under a variational Bayesian paradigm was shown to improve genomic prediction accuracy over legacy modelling approaches. This VBS-ML approach can be used to dramatically decrease the parameter burden on the network and provide a computationally feasible approach for improving genomic prediction conducted with large breeding population numbers and genetic markers.
- 2 September 2023
- Published by Springer Nature in Plant Methods
- Vol. 19 (1) , 1-14
Open AccessImproved genomic prediction using machine learning with Variational Bayesian sparsity
Background: Genomic prediction has become a powerful modelling tool for assessing line performance in plant and livestock breeding programmes. Among the genomic prediction modelling approaches, linear based models have proven to provide accurate predictions even when the number of genetic markers exceeds the number of data samples. However, breeding programmes are now compiling data from large numbers of lines and test environments for analyses, rendering these approaches computationally prohibitive. Machine learning (ML) now offers a solution to this problem through the construction of fully connected deep learning architectures and high parallelisation of the predictive task. However, the fully connected nature of these architectures immediately generates an over-parameterisation of the network that needs addressing for efficient and accurate predictions. Results: In this research we explore the use of an ML architecture governed by variational Bayesian sparsity in its initial layers that we have called VBS-ML. The use of VBS-ML provides a mechanism for feature selection of important markers linked to the trait, immediately reducing the network over-parameterisation. Selected markers then propagate to the remaining fully connected feed-forward components of the ML network to form the final genomic prediction. We illustrated the approach with four large Australian wheat breeding data sets that range from 2665 lines to 10375 lines genotyped across a large set of markers. For all data sets, the use of the VBS-ML architecture improved genomic prediction accuracy over legacy linear based modelling approaches. Conclusions: An ML architecture governed under a variational Bayesian paradigm was shown to improve genomic prediction accuracy over legacy modelling approaches. This VBS-ML approach can be used to dramatically decrease the parameter burden on the network and provide a computationally feasible approach for improving genomic prediction conducted with large breeding population numbers and genetic markers.
- Open Access
- 9 September 2021
- Published by Frontiers Media SA in Frontiers in Plant Science
- Vol. 12
Plant Variety Selection Using Interaction Classes Derived From Factor Analytic Linear Mixed Models: Models With Independent Variety Effects
A major challenge in the analysis of plant breeding multi-environment datasets is the provision of meaningful and concise information for variety selection in the presence of variety by environment interaction (VEI). This is addressed in the current paper by fitting a factor analytic linear mixed model (FALMM) then using the fundamental factor analytic parameters to define groups of environments in the dataset within which there is minimal crossover VEI, but between which there may be substantial crossover VEI. These groups are consequently called interaction classes (iClasses). Given that the environments within an iClass exhibit minimal crossover VEI, it is then valid to obtain predictions of overall variety performance (across environments) for each iClass. These predictions can then be used not only to select the best varieties within each iClass but also to match varieties in terms of their patterns of VEI across iClasses. The latter is aided with the use of a new graphical tool called an iClass Interaction Plot. The ideas are introduced in this paper within the framework of FALMMs in which the genetic effects for different varieties are assumed independent. The application to FALMMs which include information on genetic relatedness is the subject of a subsequent paper.
- 9 September 2021
- Published by Frontiers Media SA in Frontiers in Plant Science
- Vol. 12
Open AccessPlant Variety Selection Using Interaction Classes Derived From Factor Analytic Linear Mixed Models: Models With Independent Variety Effects
A major challenge in the analysis of plant breeding multi-environment datasets is the provision of meaningful and concise information for variety selection in the presence of variety by environment interaction (VEI). This is addressed in the current paper by fitting a factor analytic linear mixed model (FALMM) then using the fundamental factor analytic parameters to define groups of environments in the dataset within which there is minimal crossover VEI, but between which there may be substantial crossover VEI. These groups are consequently called interaction classes (iClasses). Given that the environments within an iClass exhibit minimal crossover VEI, it is then valid to obtain predictions of overall variety performance (across environments) for each iClass. These predictions can then be used not only to select the best varieties within each iClass but also to match varieties in terms of their patterns of VEI across iClasses. The latter is aided with the use of a new graphical tool called an iClass Interaction Plot. The ideas are introduced in this paper within the framework of FALMMs in which the genetic effects for different varieties are assumed independent. The application to FALMMs which include information on genetic relatedness is the subject of a subsequent paper.
- 6 March 2021
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 134 (5) , 1387-1407
Genetic analysis of wheat (Triticum aestivum) adaptation to heat stress
Key message Adaptation to abiotic stresses such as high-temperature conditions should be considered as its independent components of total performance and responsiveness. Abstract Understanding and identifying improved adaptation to abiotic stresses such as heat stress has been the focus of a number of studies in recent decades. However, confusing and potentially misleading terminology has made progress difficult and hard to apply within breeding programs selecting for improved adaption to heat stress conditions. This study proposes that adaption to heat stress (and other abiotic stresses) be considered as the combination of total performance and responsiveness to heat stress. In this study, 1413 doubled haploid lines from seven populations were screened through a controlled environment assay, subjecting plants to three consecutive eight hour days of an air temperature of 36 °C and a wind speed of 40 km h−1, 10 days after the end of anthesis. QTL mapping identified a total of 96 QTL for grain yield determining traits and anthesis date with nine correlating to responsiveness, 72 for total performance and 15 for anthesis date. Responsiveness QTL were found both collocated with other performance QTL as well as independently. A sound understanding of genomic regions associated with total performance and responsiveness will be important for breeders. Genomic regions of total performance, those that show higher performance that is stable under both stressed and non-stressed conditions, potentially offer significant opportunities to breeders. We propose this as a definition and selection target that has not previously been defined for heat stress adaptation.
- 6 March 2021
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 134 (5) , 1387-1407
Genetic analysis of wheat (Triticum aestivum) adaptation to heat stress
Key message Adaptation to abiotic stresses such as high-temperature conditions should be considered as its independent components of total performance and responsiveness. Abstract Understanding and identifying improved adaptation to abiotic stresses such as heat stress has been the focus of a number of studies in recent decades. However, confusing and potentially misleading terminology has made progress difficult and hard to apply within breeding programs selecting for improved adaption to heat stress conditions. This study proposes that adaption to heat stress (and other abiotic stresses) be considered as the combination of total performance and responsiveness to heat stress. In this study, 1413 doubled haploid lines from seven populations were screened through a controlled environment assay, subjecting plants to three consecutive eight hour days of an air temperature of 36 °C and a wind speed of 40 km h−1, 10 days after the end of anthesis. QTL mapping identified a total of 96 QTL for grain yield determining traits and anthesis date with nine correlating to responsiveness, 72 for total performance and 15 for anthesis date. Responsiveness QTL were found both collocated with other performance QTL as well as independently. A sound understanding of genomic regions associated with total performance and responsiveness will be important for breeders. Genomic regions of total performance, those that show higher performance that is stable under both stressed and non-stressed conditions, potentially offer significant opportunities to breeders. We propose this as a definition and selection target that has not previously been defined for heat stress adaptation.
- Open Access
- 1 September 2018
- Published by Oxford University Press (OUP) in G3: Genes, Genomes, Genetics
- Vol. 8 (9) , 2889-2899
Optimising Genomic Selection in Wheat: Effect of Marker Density, Population Size and Population Structure on Prediction Accuracy
Genomic selection applied to plant breeding enables earlier estimates of a line’s performance and significant reductions in generation interval. Several factors affecting prediction accuracy should be well understood if breeders are to harness genomic selection to its full potential. We used a panel of 10,375 bread wheat (Triticum aestivum) lines genotyped with 18,101 SNP markers to investigate the effect and interaction of training set size, population structure and marker density on genomic prediction accuracy. Through assessing the effect of training set size we showed the rate at which prediction accuracy increases is slower beyond approximately 2,000 lines. The structure of the panel was assessed via principal component analysis and K-means clustering, and its effect on prediction accuracy was examined through a novel cross-validation analysis according to the K-means clusters and breeding cohorts. Here we showed that accuracy can be improved by increasing the diversity within the training set, particularly when relatedness between training and validation sets is low. The breeding cohort analysis revealed that traits with higher selection pressure (lower allelic diversity) can be more accurately predicted by including several previous cohorts in the training set. The effect of marker density and its interaction with population structure was assessed for marker subsets containing between 100 and 17,181 markers. This analysis showed that response to increased marker density is largest when using a diverse training set to predict between poorly related material. These findings represent a significant resource for plant breeders and contribute to the collective knowledge on the optimal structure of calibration panels for genomic prediction.
- 1 September 2018
- Published by Oxford University Press (OUP) in G3: Genes, Genomes, Genetics
- Vol. 8 (9) , 2889-2899
Open AccessOptimising Genomic Selection in Wheat: Effect of Marker Density, Population Size and Population Structure on Prediction Accuracy
Genomic selection applied to plant breeding enables earlier estimates of a line’s performance and significant reductions in generation interval. Several factors affecting prediction accuracy should be well understood if breeders are to harness genomic selection to its full potential. We used a panel of 10,375 bread wheat (Triticum aestivum) lines genotyped with 18,101 SNP markers to investigate the effect and interaction of training set size, population structure and marker density on genomic prediction accuracy. Through assessing the effect of training set size we showed the rate at which prediction accuracy increases is slower beyond approximately 2,000 lines. The structure of the panel was assessed via principal component analysis and K-means clustering, and its effect on prediction accuracy was examined through a novel cross-validation analysis according to the K-means clusters and breeding cohorts. Here we showed that accuracy can be improved by increasing the diversity within the training set, particularly when relatedness between training and validation sets is low. The breeding cohort analysis revealed that traits with higher selection pressure (lower allelic diversity) can be more accurately predicted by including several previous cohorts in the training set. The effect of marker density and its interaction with population structure was assessed for marker subsets containing between 100 and 17,181 markers. This analysis showed that response to increased marker density is largest when using a diverse training set to predict between poorly related material. These findings represent a significant resource for plant breeders and contribute to the collective knowledge on the optimal structure of calibration panels for genomic prediction.
- Open Access
- 8 September 2017
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 130 (12) , 2543-2555
Increased genomic prediction accuracy in wheat breeding using a large Australian panel
Key message Genomic prediction accuracy within a large panel was found to be substantially higher than that previously observed in smaller populations, and also higher than QTL-based prediction. Abstract In recent years, genomic selection for wheat breeding has been widely studied, but this has typically been restricted to population sizes under 1000 individuals. To assess its efficacy in germplasm representative of commercial breeding programmes, we used a panel of 10,375 Australian wheat breeding lines to investigate the accuracy of genomic prediction for grain yield, physical grain quality and other physiological traits. To achieve this, the complete panel was phenotyped in a dedicated field trial and genotyped using a custom AxiomTM Affymetrix SNP array. A high-quality consensus map was also constructed, allowing the linkage disequilibrium present in the germplasm to be investigated. Using the complete SNP array, genomic prediction accuracies were found to be substantially higher than those previously observed in smaller populations and also more accurate compared to prediction approaches using a finite number of selected quantitative trait loci. Multi-trait genetic correlations were also assessed at an additive and residual genetic level, identifying a negative genetic correlation between grain yield and protein as well as a positive genetic correlation between grain size and test weight.
- 8 September 2017
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 130 (12) , 2543-2555
Open AccessIncreased genomic prediction accuracy in wheat breeding using a large Australian panel
Key message Genomic prediction accuracy within a large panel was found to be substantially higher than that previously observed in smaller populations, and also higher than QTL-based prediction. Abstract In recent years, genomic selection for wheat breeding has been widely studied, but this has typically been restricted to population sizes under 1000 individuals. To assess its efficacy in germplasm representative of commercial breeding programmes, we used a panel of 10,375 Australian wheat breeding lines to investigate the accuracy of genomic prediction for grain yield, physical grain quality and other physiological traits. To achieve this, the complete panel was phenotyped in a dedicated field trial and genotyped using a custom AxiomTM Affymetrix SNP array. A high-quality consensus map was also constructed, allowing the linkage disequilibrium present in the germplasm to be investigated. Using the complete SNP array, genomic prediction accuracies were found to be substantially higher than those previously observed in smaller populations and also more accurate compared to prediction approaches using a finite number of selected quantitative trait loci. Multi-trait genetic correlations were also assessed at an additive and residual genetic level, identifying a negative genetic correlation between grain yield and protein as well as a positive genetic correlation between grain size and test weight.