BGRI 2021 Virtual Technical Workshop Program

Biography: Professor Chris Gilligan is currently director of research and head of the Epidemiology and Modelling Group at the University of Cambridge. Together with his research group and colleagues, Gilligan has developed an extensive epidemiological toolkit with applications in plant and animal disease, including Covid-19. His current research is focused on the development and use of models to predict spread and identify effective strategies for the control of crop disease at scales ranging from on-farm to the landscape, regional and continental scales. Current applications involve models to inform the surveillance and control of crop disease and pests in sub-Saharan Africa (wheat rusts, cassava and banana disease and desert locust), Australia (banana and avocado disease) and India (aflatoxins on maize), Nepal and Bangladesh (wheat rusts) the US (citrus disease) and UK (range of crop and diseases of natural vegetation).

He has published more than 250 research papers and reviews. He has held a UK National Research Council Professorial Fellowship and a Royal Society Leverhulme Trust Senior Research Fellowship, and is an Honorary Fellow of the American Phytopathological Society and former President of the British Society for Plant Pathology. He was awarded a CBE in the Queen’s Birthday honours in 2015 for services to plant health in the field of epidemiology.

Abstract forthcoming

Wheat Blast in Zambia

Wheat blast caused by Magnaporthe oryzae pathotype Triticum is a destructive disease of wheat particularly in warmer and humid environments. It causes significant yield losses and poor grain quality with low test weight. In Zambia, wheat blast symptoms were first observed during 2017/2018 rainy season making it the first report in Zambia and Africa. The disease occurred again in 2018/2019 season but severity was low due to unfavourable (hot and dry) climatic conditions for disease proliferation. During 2019/2020 rainy season, severity was high as conditions were favourable for disease development. Higher disease severity and yield loss was observed in Coucal and Kwale, indicating inadequate levels of resistance to blast in these local varieties. Wheat blast is a new disease in Zambia, continued effort to identify resistant sources to be used in breeding programme to combat the deadly disease is essential. Currently, screening wheat germplasm to identify resistant genotypes is underway.

Wheat Blast: epidemiological lessons learned in South America

Wheat blast, caused by Magnaporthe oryzae Triticum pathotype (MoT), has caused widespread losses in South America and South Asia. The recent report in Zambia, Africa, demonstrates the threat of the global spread of this intractable and dangerous disease. Overall, wheat blast is an understudied disease, and a lack of sound frameworks to understand wheat blast epidemics has made its management a challenging task. Results obtained after more than a decade of working on critical knowledge gaps in wheat blast epidemiology will be shared. Results will include our understanding of wheat blast epidemics from experiments conducted in Bolivia and Brazil under field epidemic conditions. 

Early warning requires immediate or short-term actions and alerts (1 day- few month) e.g. an effective system for collecting disease surveillance and sampling data, fast and reliable characterization of pathogen samples and effective coordinated communication and dissemination of results. Early warning may also include longer term actions (6-18 months) e.g. assessment of the epidemic potential of new emerging races, accelerated breeding for resistance, adaptation of IPM based prevention and control strategies and dissemination of synthesised and integrated results via stakeholder networks and academia articles, workshops and conferences. This presentations will present “what did we learn” from new web based early warning tools developed in the frame of AGG and RustWatch. For early season surveillance of rust diseases in Europe we developed a smartphone crowdsource App and an associated Dashboard. Data indicating the local disease pressure from approximately 100 official variety testing trial sites (VCU) were integrated with disease survey data on the same maps and charts. Trap nurseries defined by RustWatch and laid out at the VCU trial sites effectively contributed to “hunting the new” races in Europe. Stem rust is spreading into new regions in Europe (south to north) and results from a network of field nurseries indicated that a majority of cultivars grown in the North are susceptible to SR races currently present in the South. A Field Nursery Data Management System manage all data, analyse and visualise data in collaboration research partners and 12 breeding companies in Europe. In the Frame of AGG we further developed the Vulnerability Mapping Tool for YR and SR races in Ethiopia. A new YR race with the preliminary name PstS16 moved from Central Asia to Ethiopia and now covers approximately half of the population. The web-tool indicate the distribution of varieties in the country, and which areas are vulnerable for severe epidemics in the up-coming main growing season. The presentation will discuss how such integrated web tools effectively can support early warning at all stages and our experiences so far.

A new approach for stakeholder driven wheat rust early-warning was launched in Europe in 2018. The initiative is based on intensified collaboration between partners within plant pathology, breeding, agronomy, agro-chemistry, molecular genetics and Information and Communication Technologies (ICT). A core activity is to provide a better understanding of pathogen biology and drivers by faster and more efficient diagnostic methods to detect new rust races and genotypes, to use them to investigate population genetic changes in Europe, to study the role of pathogen aggressiveness and temperature adaptation as epidemic drivers and to study the role of Berberis spp. as alternate host for rust fungi in Europe. This has been possible by strengthening the collaboration among European rust diagnostic laboratories, e.g., new procedures for alignment of experimental procedures and interpretation of results, and exchange and compilation of genotypic and phenotypic data using the facilities provided by the WheatRustToolbox. Disease prevention by host resistance is another focus area, providing new knowledge about rust resistance specificities in European wheat germplasm to all three wheat rust fungi, e.g., by integrating epidemiological assays (in vivo) and SNP genotyping for a large number of rust resistance genes. This also involves new approaches for off-season assessment of host resistance using new improved laboratory inoculation and microscopy techniques, and analyses of the impact of resistance gene diversification for preventing rust epidemics at field and landscape scales. The integration with existing stakeholder networks has a high priority, e.g. to increase the expertise and efficiency in rust monitoring within plant breeding, agrochemical industry, agricultural advisory services, and Value for Cultivation and Use (VCU) approval in cereals. The early warning system is implemented in five case study regions, which support development and validation of the systems in collaboration with a wide array of stakeholder groups. Agricultural advisers are involved in disease surveillance and rust sampling for later diagnosis, and giving feed-back about potential barriers in adopting the system. The case study approach will allow us to investigate the threats of invasive rust races at the regional scale and suggest IPM-based prevention and control options adjusted to the relevant agronomical practices and environmental conditions in the considered region. The Wheat Rust Toolbox plays a decisive role for storage, analyses and dissemination of results from the project, which is based on database driven information systems, web tools and mobile platforms. Data management is in accordance with the H2020 Guidelines on FAIR Data Management. RustWatch is funded by the European Union’s Horizon 2020 research and innovation programme (grant agreement 773311), engaging 12 universities/research institutes, 5 agricultural advisory services, and 7 SMEs/industries in Europe.

Biography: Nozomi Kawarazuka is a social anthropologist at the International Potato Center, based in Hanoi, Vietnam. She conducts a wide range of interdisciplinary agricultural research, such as understanding the social and gender dimensions of technology adoption, labour and cropping systems, and intersectional constraints of women, youth and ethnic minorities in agriculture. Her research also draws on political economy and political ecology from gender perspectives to understand the processes of marginalization and exclusion in rapidly changing urban and rural food systems. Her current regional focus is Southeast and South Asia. She has a Ph.D. from University of East Anglia, UK.

The climate crisis calls for fast and accurate development of improved crop varieties with local adaptation. In smallholder cropping systems worldwide, an estimated 500 million farmers support the livelihoods of up to 2 billion people. Here, heterogenous cropping environments with limited buffering capacity may lead to uneven agronomic performance of varieties currently produced by breeding. In order to fully harness the potential of breeding driven by genomic selection (GS) in smallholder farming settings, genetic diversity data must be coupled with phenotypic data that is truly representative of farmer fields. Here, we present a data-driven decentralized breeding approach called 3D-breeding that leverages crowd-sourced observations of crop performance in farmer fields to enhance the accuracy of GS. 3D-breeding uses a ranking order approach based on a Plackett-Luce model to estimate the worth of genotypes and efficiently predict performance in unobserved farmer fields. We developed 3D-breeding using the case study of durum wheat in the Ethiopian highlands, collaborating with hundreds of smallholder farmers who are expert in wheat cultivation. We selected and genotyped a collection 400 durum wheat landraces and improved varieties representing the agrobiodiversity of Ethiopian durum wheat. The collection was characterized for agronomic performance in experimental stations in Ethiopia following a centralized breeding scheme, then a subset of the collection was tested in 3D-breeding to bring the selection in in farmer fields. Each of 1,165 smallholder farmers across Ethiopia was asked to grow a different combination of four varieties from the subset in their own fields. At the end of the growing season, we collected farmers’ decentralized preferences and we measured grain yield in their plots. We tested the prediction accuracy of wheat performance achieved by 3D-breeding against a benchmark representing a centralized GS approach trained on station data. We found that 3D-breeding could double the accuracy of GS for durum wheat yield in smallholder farmer fields. We tested 3D-breeding against different competitive GS scenarios, showing that our method maintained an edge for prediction accuracy also for farmers’ appreciation, a major driver of varietal uptake. 3D-breeding could consistently predict yield and farmers’ appreciation across different seasons and, if used for varietal selection, would prioritize different wheat genotypes from those currently recommended for the area. The advantage of 3D-breeding lies in capturing the broad diversity of farmer fields and end user needs in a big data dimension, providing a better representation of the local growing conditions that are targeted by breeding programs. We discuss the value of our decentralized method beyond the durum wheat case study as a cost-effective and efficient complement to breeding for challenging cropping environments.

Seed is the most crucial input in crop production and serves as a prime channel for technology (variety) dissemination. Despite rigorous efforts to promote the modern wheat varieties in India, the seed and varietal replacement rates are low, barring a few regions like Punjab and Haryana. To counter this, inter alia, frontline demonstration (FLD) – an extension method wherein the recent varieties (treatment) are demonstrated at the farmers’ field along with their preferred variety (control) in the same piece of land for evaluation – has been widely advocated. The purpose of the FLD is to increase the awareness of recent varieties for harnessing the yield potential. It also gives an opportunity to the beneficiaries in understanding the complete package of practices involved with the demonstrated technology under the guidance of experts that gets reflected on the profitability and welfare of the farmers. In the milieu, we evaluated the impact of the recently released wheat varieties (including rust resistant, biofortified, saline tolerant, etc.) demonstrated at the farmers’ field in India. Using an experimental approach, the seeds of the newly released 38 wheat varieties suited to different production conditions were offered and demonstrated during 2019-20 crop season at randomly selected 1607 farmers’ fields across India covering 19 states from 5 wheat growing agro-climatic zones. Analysis of the yield gap, using the survey data from the 85 cooperating centers, indicated a significant difference ranging from 5.51% to 59.52% between the demonstrated plot (treatment: modern variety) vis-à-vis check plot (control: farmers’ choice of variety). The level of yield gap was higher in the hilly and eastern regions of India. Among the demonstrated varieties, ‘Hybrid Indore 8759’ (short as HI 8759), a durum wheat variety demonstrated in Madhya Pradesh (Central Zone) registered the highest yield level (7700 kg/ha) with an estimated yield gain of 32.76% over the control. The gross margin analysis supported the positive effect of demonstrations with an incremental benefit estimated at ₹ 11687 (~158US$) per hectare. We also fitted a profit function (regression analysis) to capture the impact of the demonstration using wheat-growing zones as a dummy variable along with other socio-economic factors. The results indicated that the profit levels are influenced by the market price of the harvested produce, and the region wherein the variety is demonstrated. The cost of variable inputs has been a major constraint in the adoption of the technology. Clearly, our analysis affirms the positive impact of the demonstration at the farmers’ field. Such demonstrations have to be continued for improving the socio-economic status and livelihood of the farmers, especially the vulnerable groups and include disadvantaged regions like northeast India wherein the recent varieties adoption is substantially low.

Tan spot, caused by Pyrenophora tritici-repentis (Ptr), is a major foliar disease of wheat. A collection of Ptr isolates from Tunisia, one of the main secondary centers of diversification of durum wheat, was tested for phenotypic race classification based on virulence on a host differential set consisting of four hexaploid wheats (‘Glenlea’, 6B662, 6B365 and ‘Salamouni’), and for the presence of the necrotrophic effector (NE) genes ToxA, ToxB, and toxb by PCR analysis. While races 2, 4, 5, 6, 7, and 8 were identified according to their virulence phenotypes, with races 5 and 7 predominant, PCR testing indicated the presence of ‘atypical’ isolates that induced necrosis on the wheat differential ‘Glenlea’, but lacked the expected ToxA gene. This suggested the involvement of other NEs in the Ptr/wheat interaction responsible for symptom development. Genetic diversity and the Ptr population structure were explored further through the evaluation of 59 Tunisian isolates and 35 isolates from Algeria, Azerbaijan, Canada, Iran, and Syria using 24 simple sequence repeat markers. Average genetic diversity, overall gene flow and percentage polymorphic loci were estimated as 0.58, 2.09, and 87%, respectively. Analysis of molecular variance showed that 81% of the genetic variance occurred within populations and 19% between populations. Cluster analysis by the unweighted pair group method indicated that ToxB- isolates grouped together and were distantly related to ToxB+ isolates. Based on Nei’s analysis, the global collection clustered into two distinct groups according to their region of origin; the Algerian, Azerbaijani, Canadian, and Iranian populations of Ptr clustered together and separated from the Tunisian and Syrian isolates. The results suggest that both geographic origin and the host-specificity imposed by different NEs can lead to differentiation among Ptr populations.

Biography: Alison Bentley is the director of CIMMYT’s Global Wheat Program and the CGIAR Research Program on Wheat (WHEAT). She leads and manages a team of 40 international scientists who use scientific approaches to develop improved wheat germplasm. This germplasm is distributed to around 200 cooperators in wheat-producing countries worldwide, and is responsible for the derived varieties being grown on more than 50% of the spring wheat area in developing countries.

Prior to joining CIMMYT in November 2020, she was working in the UK focused on translation of fundamental scientific breakthroughs into tangible impacts for the agri-food sector. Her research combines genetics and genomics to develop and deliver new tools and technology to improve plant breeding and crop production. She has a doctorate in Agricultural Science and PhD in Agriculture from The University of Sydney, Australia.

Multi-environment trials are useful resources to evaluate breeding lines and achieve more accurate genomic prediction and quantitative trait loci (QTL) detection. Such resources can be best exploited with models that fit genotype-by-environment interactions (GxE) and integrating different data sources. Here, we developed a mixed model methodology for the incorporation of environmental covariates (EC, e.g. rainfall, humidity, temperature) directly into genomic prediction. We validated the new model in a large wheat population (>3000 lines) being bred for heat tolerance and evaluated across multiple years and locations. Lines were deeply phenotyped for in-field agronomic and phenomic (e.g. thermal imaging), seed quality, resistance to three rust diseases and 13 seed micronutrient traits. Two sowing dates (normal and late) were used to test heat susceptibility.  ECs were reduced via principal component analysis to a much smaller set explaining the most environmental variation. GxEC modelling increased genomic prediction accuracy for yield by up to 19%. Further, we ran a meta-genome wide association analysis on the same data to detect common QTL with consistent effect across multiple environments with normal and heat stress conditions. Several loci that contribute to the accumulation of micronutrient elements were also detected. The germplasm resource, methods developed, and QTL detected will increase our understanding to the interaction of wheat cultivars with heat stress and allow for the genomic prediction of performance for projected future climate scenarios to pre-breed more climate resilient crops.

In any field of science, Occam’s razor can apply: “entities should not be multiplied beyond necessity”. So it is that any infection probability model’s goal is also to get the most accurate results with the simplest model method. With the experience of maintaining near real-time wheat rust early warning systems in developing countries, we understand that a perfect suitability model is unachievable, but we try our best to create the most effective model we can in the given circumstances. It is inevitable that compromises must be made and one has to distinguish what is the acceptable quality from a model method and what is too much effort or cost of computation. So what is too simple and what is too complex? Here we share our thoughts and results on different methods and their computational cost. We also give advice on our approach to build an environmental suitability model.

Data-driven selection decisions play a pivotal role in enhancing the efficiency of breeding programs, as they favor improved breeding outcomes and optimization of current strategies. Hence, the CIMMYT spring bread wheat breeding program has adopted a holistic data-driven selection approach that leverages phenotypic data, genomic-estimated breeding values (GEBVs) and selection indices for precise selection of lines and parents in different stages of the breeding cycle. In stages 1 and 2 of yield testing, routine integration of phenotypic selection with genomic selection using the GEBVs from both within cycle and previous cycle training sets, has facilitated increasing the selection accuracy for grain yield. In addition, for traits like drought tolerance, heat tolerance, grain zinc, grain iron, stripe rust, Septoria tritici blotch, spot blotch, Fusarium head blight and blast that are not phenotyped in the stage 1 lines, a powerful selection approach combining genomic selection and selection index has been successfully deployed to obtain the GEBVs and the net genetic merit of genotypes for line advancement and parental selection. Furthermore, to accelerate the rate of genetic gain for grain yield in CIMMYT’s target sites in South Asia, grain yield GEBVs are delivered to national partners and utilized in targeting lines for the different target population of environments. All these data-driven strategies have enabled efficient selection in the CIMMYT wheat breeding program and continued efforts are taken to optimize the application of genomic selection and selection indices for maximizing the expected genetic gain.

How to Play BGRI Trivia !

Get Kahoot!

1) Use your phone — or a separate device from the workshop Zoom (eg., tablet, different computer).

2) Download Kahoot app from Apple or Play store.
OR go to kahoot.it when we announce the game.

3) Enter Game Pin on your phone when it is displayed on the workshop Zoom screen.

4) Enter your Screen Name.  It can be a fake name!

Play TRIVIA

• Look on the workshop Zoom screen for the TRIVIA question.
• Enter your answer on your phone.

Good Luck!!

Borlaug Global Rust Initiative Vice-chair Ronnie Coffman will present a Lifetime Achievement Award and deliver closing remarks for the 2021 Virtual Technical Workshop.