Developing enhanced breeding methodologies for oats for human health and nutrition

Lead Research Organisation: National Inst of Agricultural Botany
Department Name: Centre for Research


The central objective of this proposal is to apply state of the art high throughput breeding and phenotyping approaches to the genetic improvement of oats, focusing on yield, and grain and milling quality, key targets for the economic sustainability of the crop and for the milling industry. The project addresses some of the major challenges facing UK agriculture in terms of the sustainable production of safe and nutritous food.

The overall aim of this LINK project is to incorporate high throughput approaches to the IBERS oat breeding programme, to develop strategies to improve yield and other targets ranked as priorities by our industrial partners which are currently difficult or impossible to select for at early stages of breeding cycles. Marker assisted selection (MAS) represents one route to achieve this. It has been successful for introgression of major traits controlled by one or a few genes of large effect, but is difficult with more complex traits governed by many genes, each with a small effect. MAS is used in the IBERS oat breeding programme, largely based on predictions derived from a few markers linked to large effect quantitative trait loci (QTL). Association mapping (AM) will be used to identify further marker-trait associations enabling rapid selection or introgression within the breeding programme. In this project, genomic selection (GS) will be applied to a range of traits, and selections will be validated by comparison with breeder and conventional marker assisted (MAS) selections. Increasingly complex models will be developed in the course of the programme, and an accelerated breeding cycle driven by GS and MAS will be initiated. Traits which may predict yield will be identified by detailed phenomic and field trial analysis of a model winter oat population and an association genetics panel of advanced breeding lines. Metabolic profiling and micro-scale analytical methods will be used to develop further predictive screens. Chip-based high throughput genotyping will be used to predict breeding values; genotype and phenotype data will be incorporated into a pedigree database to further facilitate 'intelligent' breeding design. The existing Illumina 6K iSelect bead assay will be expanded to include SNPs identified from UK winter and European germplasm which have significantly different genetic bases from the bulk of varieties used to develop the initial assay set. Genotyping by Sequencing will become the main platform by the end of the project to take advantage of expected sequence throughput improvements.

This project proposal addresses sustainable agricultural production at the interface of two BBSRC strategic priority areas: crop science and healthy and safe food. It is of high strategic relevance, specifically in enhancing crop productivity and quality, enhanced nutritional composition, increasing sustainability of crop production and understanding and exploiting genomics and the genetic diversity in plants (crop science). It will also investigate the potential of novel nutrient supplies from plants (healthy and safe food).

This proposal is being submitted through the BBSRC stand-alone LINK scheme. The project will benefit from the involvement of the major oat variety development company in the UK (Senova) and the British Oat and Barley Millers Association (BOBMA) representing the major oat milling companies within the UK such as PepsiCo/Quaker, Morning Foods, European Oat Millers, Grampian Oats, Hogarths and SpeediCook. Involvement of industrial partners will allow for identification and review of key targets and delivery of the outcomes of this project alongside the academic partners.

Technical Summary

The demand for high quality food grade oats is increasing annually, driven by its proven health benefits and by product development by the milling and cereals industries. Despite an expanding market, the oat crop is facing increasing competition from other arable crops and that is impacting on the UK oat area grown. This project will apply the latest genetic tools and resources, including genomic selection, to improve key traits that will increase the production and utilisation of oats, and to enhance grain yield, quality and composition. These are increasingly important traits for the UK oat milling industry to fully exploit the nutritional characteristics of the oat grain and key economic drivers for product development. Optimal strategies for high throughput phenotyping and genotyping will be developed. We will establish an efficient high throughput genotyping platform for use across the breeding programme. The advantages of genomic selection over conventional breeding approaches for defined traits will be determined and practical limits established for the use of GS on a wider scale. It will use a range of unique genetic material to dissect the genetic and environmental factors contributing to yield variation and use that information in an innovative oat breeding programme. Breeding programme populations will be used to extend the range of available marker-trait assays to improve selection efficiency in early generations and ensure uniformity in late generations. Phenotyping of key traits concerned with yield, milling quality and with human health and nutrition will be conducted and the impact of changing grain quality characters (beta-glucan or starch content/type) determined. GS results will be validated and gain compared with conventional and MAS approaches on the same populations. This will dramatically increase the efficiency of breeding new varieties and identify regions of the genome associated with key traits through detailed stakeholder discussions.

Planned Impact

The major beneficiaries of this research will be:
Plant breeding community: Information on the application of genomic selection in an oat breeding programme will be of value to other breeders of oats and other cereal crops. Development of high throughput screens for key traits associated with milling quality, grain composition and yield that can be used at the early stages of the breeding programme necessitating lower seed quanities will enhance selection ofr yield and quality and wil be of direct benefit to genomic selection approaches but more generally would be beneficial to other breeding programmes.

Arable sector: The project will have a significant economic and environmental benefit on the arable sector. Oats are recognised as a low input cereal crop, that can be grown in arable rotations across the UK. It is an excellent break crop and therefore has a major role in sustainable arable rotations. Application of genomic selection and high throughput phenotyping to the breeding of improved oat varieties with greater yield and grain and milling quality that meets the needs of end-users and the oat value-chains will improve the economics of growing the crop and its value to the arable sector. This will ensure that oats remain a competitive crop for arable production and that the benefit of growing oats in arable rotations is realised.

Milling industry:Increasing awareness of the health benefits of eating oats, largely due to the beta-glucan content of the grain but also to other grain compositinal characteristics, is increasing the demand for oats and oat based products at a time when the oat crop is increasingly challenged by competition from other cereals. using GS approaches to increase the yield of new oat varieties will increase the financial return of growing oats and will help to ensure that the milling industry has a greater security of supply. The focus on grain quality and composition will also ensure that new varieties have the essential grain quality characteristics that ensure the economic efficiency of the mills and cost benefit. The milling industry will also benefit from the focus on grain composition which will enhance the potential of using oats for different end-uses.

Society: Direct benefit will be gained from ensuring the supply of a cereal with recognised health benefits. Enhanced yield and quality of oat varieties will improve the economics of growing oat crops and so ensure that oats, as alow input cereal remain a part of sustainable arable rotations, increasing cropping diversity.


10 25 50
Description Test simulations in wheat, while awaiting the delivery of data in oats, have confirmed that accurate trait predictions within a biparental cross are possible with very small numbers of markers and lines.

Methods to increase the accuracy of trait predictions by combining information across multiple sources of information have been developed and published.

Using data from an oats mapping population, we have shown that predictions of important agronomic traits in oats, like plant height, flowering time and seed yield are possible with the accuracy required for use in breeding.

Work has been extended to give more detailed predictions of F6 inbred lines from F2:F4 .families which will be useful in breeding oats and other cereals. We are currently preparing a manuscript describing this for publication and have shared the information with all project partners

2019 - the manuscript describing our work has now been accepted for publication: Mellers et al (in press) "Implementing withincross genomic prediction to reduce oat breeding costs". This will be published in the next reporting year. A final report of our findings has also been shared with project partners.
Exploitation Route Integration of multiple trait prediction into the line evaluation process needs to be evaluated. Prediction over greater genetic distances needs to be tested.

We need to work out how best to modify the breeding programme to exploit the ability to predict traits within crosses.

Used to predict yield and other traits on inbred lines derived from a biparental cross from data collected earlier in the breeding process.
Sectors Agriculture, Food and Drink

Description We have shown that various traits of agronomic importance can be predicted within crosses of oats and have shown how this can be incorporated into breeding programmes at low-cost in Mellers et al (in press) "Implementing withincross genomic prediction to reduce oat breeding costs".
First Year Of Impact 2019
Sector Agriculture, Food and Drink
Impact Types Economic

Title DiPR: Differentially Penalized Regression 
Description A simple method of independently weighting two or more sets of variables in ridge regression, while using standard ridge regression software. Useful in genomic prediction models with separate classes of genetic markers, and for incorporating different sets of predictors eg plant metabolites and genetic markers. 
Type Of Material Data analysis technique 
Year Produced 2015 
Provided To Others? Yes  
Impact Is being tested in various genomic selection projects. 
Description Collaboration with Portugal 
Organisation National Institute of Agrarian and Veterinary Research
Country Portugal 
Sector Public 
PI Contribution I advise the Portuguese FASTBREED project: Implementation of a bread and durum wheat breeding program based on genomic selection using genomic selection. This is a project to introduce genomic selection into the state wheat breeding programmes. In 2016, at the kick off meetings, I gave two seminars; one to the breeders and one to molecular biologists and bioinformaticians to explain how genomic selection works. I also proposed a crossing and testing programme with their own breeding material to introduce genomic selection. This is now being put in place.
Collaborator Contribution Portuguese breeders and geneticists are running the breeding programme and carrying out all genotyping, analytical and other activities.
Impact None so far.
Start Year 2016
Title Diferentially Penalized Regression (DiPR) 
Description A simple method of using existing software for ridge regression to independently penalized two sets of predictors. THis is never less accurate than using either set alone or both sets pooled and treated as one. Applications are predicting traits from a combination of genetic markers and metabolites or predicting hybrid performance from co-dominant markers, fitting separate additive an dominance effects at each. Can be used in genomic selection and association mapping. 
Type Of Technology Software 
Year Produced 2014 
Open Source License? Yes  
Impact Prediction of wheat yields and other agronomic traits from seed metabolites and genetic markers. Combining known markers for QTL with background markers in genomic selection / prediction in an optimal manner. Used in two publications: Ward J, Rakszegi M, Bedo S, Shewry PR, Mackay I (2015) Differentially penalized regression to predict agronomic traits from metabolites and markers. BMC Genetics 16:19 Bentley AR, Scutari M, Gosman N, Faure S, Bedford F, Howell P, Cockram J, Rose GA, Barber T, Irigoyen J, Horsnell R, Pumfrey C, Winnie E, Schacht J, Beauchêne K, Praud S, Greenland A, Balding D, Mackay I J (2014) Applying association mapping and genomic selection to the dissection of key traits in elite European wheat. Theoretical and Applied Genetics, 127:2619-2633 
Description Camb. Plant Sci epidemiology groups 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Professional Practitioners
Results and Impact A seminar to the two epidemiology groups in the University of Cambridge Plant Sciences Department. To introduce to the methods and approaches of trait mapping and genomic selection in crops, highlighting some overlaps with epidemiological methodology.
Year(s) Of Engagement Activity 2016
Description JHI offsite meeting 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Gave talk to James Hutton Institute scientific staff at their annual off-site workshop in Dunkeld on genomic selection and how it can be implemented in small breeding programmes.
Year(s) Of Engagement Activity 2016
Description Landmark artucle on GS 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact I wrote an article on genomic selection "Genomic selection and the quantitative geneticists' revenge" for the NIAB members quarterly journal: LandMark Bulletin. This described the potential of genomic selection in plant breeding in lay terms.
Year(s) Of Engagement Activity 2016
Description QMPB 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact NIAB run a two week intensive course in Quantitative Methods for Plant Breeding, covering relevant aspects of statistics, quantitative genuetics and population genetics. There are 25 participants every year. To date, no-one has said they would not recommend the course to others. The course has also been put on in Australia, France, India (once at ICRISAT and once at the Punjab Agricultural University) and Malaysia. We update it every year to disseminate developments and methods resulting from out own research work in MAGIC, genomic selection, association mapping and plant breeding strategy. It has resulted in a global set of new contacts, some of which have resulted in successful collaborations.
Year(s) Of Engagement Activity 2008,2009,2010,2011,2012,2013,2014,2015,2016
Description Turkey Wheat Workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Participation in one week workshop over multiple sites in Turkey to discuss UK - Turkish collaboration over wheat research, representing the NIAB Genetics and Breeding group on behalf of Dr Alison Bentley, the Director of Genetics and Breeding. Aside from questions and discussions at the time, it resulted in an agreement to sent some of our most resistant yellow rust lines to Turkey for screening, a novel approach to participatory plant breeding incorporating genomic selection and a collaborative grant application to the GCRF.
Year(s) Of Engagement Activity 2016