The development of new cultivars better adapted to specific growing conditions is a key strategy to meet an ever-increasing growing global food demand and search for more sustainable cropping systems. This is even more crucial in the context of a changing climate. Ecophysiological models and advanced computational techniques (e.g., sensitivity analysis, SA) represent powerful tools to analyze genotype (G) by environment (E) interactions, thus supporting breeders in identifying key traits for specific agro-environmental contexts. However, limits for the effective use of mathematical models within breeding programs are represented by the uncertainty in the distribution of plant trait values, the lack of processes dealing with resistance/tolerance traits in most ideotyping studies, the partial suitability of current crop models for ideotyping purposes, and the absence of modelling tools directly usable by breeders. The aim of this research was to address these issues improving methodologies already in use, proposing new paradigms for the development of crop models explicitly targeting ideotyping applications and developing tools that would encourage a deep interaction of the modelling and breeding communities. The focus was on rice, for its role as staple food for more than a half of world’s population, and on resistance/tolerance traits to biotic/abiotic stressors, for their central role in increasing crop adaptation. Moreover, current conditions and climate change projections were considered, to support the definition of strategies for breeding in the medium-long term. A standard procedure to quantify − and manage − the impact of the uncertainty in the distribution of plant trait values was developed, using the WARM rice model and the Sobol’ method as case study. The approach is based on a SA (generating sample of parameter distributions) of a SA (generating samples of parameters for each generated distribution) using distributions of jackknife statistics calculated on literature values to reproduce the uncertainty in defining parameters distributions. As a practical implication, the procedure developed allows identifying plant traits whose uncertainty in distribution can alter ideotyping results, i.e., traits whose distributions could need to be refined. Global SA was then used to identify rice traits putatively producing the largest yield benefits in five contrasting districts in the Philippines, India, China, Japan and Italy. The analysis involved phenotypic traits dealing with light interception, photosynthetic efficiency, tolerance to abiotic stressors, resistance to fungal pathogens and grain quality. Results suggested that breeding for traits involved with disease resistance and tolerance to cold- and heat-induced spikelet sterility could provide benefits similar to those obtained from improving traits affecting potential yield. Instead, advantages resulting from varying traits involved with grain quality were markedly frustrated by inter-annual weather variability. Since results highlighted strong G×E interactions, a new index to derive district-specific ideotypes was developed. Given the key role of biotic/abiotic stressors in determining actual yield and the deep impact of related G×E interactions, a study was carried out by explicitly focusing on the definition of rice ideotypes improved for their resistance to fungal pathogens and tolerance to abiotic constraints (temperature shocks inducing sterility). The analysis was carried out at district level with a high spatial resolution (5 km × 5 km elementary simulation unit), targeting the improvement of the most representative 34 varieties in six Italian rice districts. Genetic improvement was simulated via the introgression of traits from donor varieties. Results clearly showed that breeders should focus on increasing resistance to blast disease, as this appears as a factor markedly limiting rice yields in Italy, regardless of the districts and climate scenarios, whereas benefits deriving from improving tolerance to cold-induced sterility could be markedly affected by G×E interactions. To reduce the risk of discrepancies between in silico ideotypes and their in vivo realizations, both studies involved only model parameters with a close relationship with phenotypic traits breeders are working on. However, a long-term strategy to overcome limitations related with the partial suitability of available models would be building new ideotyping-specific models explicitly around traits involved in breeding programs. This proposal for a paradigm shift in model development was illustrated taking salt stress tolerance and rice as a case study. Dedicated growth chamber experiments were conducted to develop a new model explicitly accounting for tolerance traits modulating Na+ uptake and distribution in plant tissues, as well as the impact of the accumulated Na+ on photosynthesis, senescence and spikelet sterility. An ideotyping study was conducted at two sites (in Greece and California) characterized by different seasonal dynamics of salinity in field water. Results showed how, under different scenarios, traits assuring the largest contribution to the overall tolerance could refer to completely different physiological mechanisms: tissue tolerance in one case, sodium exclusion in the other. This encourages the development of explicit trait-based approaches to increase the integration of crop models within breeding programs. A parallel path to achieve this goal is the development of modelling platforms targeting breeders as final users, who does not have necessarily in-depth skills in crop modelling and IT. The platform ISIde, derived from a close collaboration between target users, biophysical modelers and IT specialists, represents the first prototype of a platform specifically developed for being used directly by breeders to evaluate in silico improved varieties at district level. This thesis demonstrated the usefulness of simulation models for the definition of ideotypes for specific agro-environmental conditions. Targeting ideotyping applications, new methodologies, paradigms for model development and modelling tools were developed, thus contributing to improve the potential of crop modelling to support breeding programs. Future developments will target researches aimed at overcoming the limits behind this study, i.e., (i) absence of explicit interactions between traits, (ii) no adaptation strategies considered, and (iii) lack of approaches for the simulation of the evolutionary potential of pathogens in response to long-term climate variations and increased host resistance.
IN SILICO IDEOTYPING: DEFINITION AND EVALUATION OF RICE IDEOTYPES IMPROVED FOR RESISTANCE/TOLERANCE TRAITS TO BIOTIC AND ABIOTIC STRESSORS UNDER CLIMATE CHANGE SCENARIOS / L. Paleari ; supervisor: R. Confalonieri ; coordinatore scuola di Dottorato: D. Bassi. - Milano : Università degli studi di Milano. DIPARTIMENTO DI ECONOMIA, MANAGEMENT E METODI QUANTITATIVI, 2017 Mar 17. ((29. ciclo, Anno Accademico 2016.
|Titolo:||IN SILICO IDEOTYPING: DEFINITION AND EVALUATION OF RICE IDEOTYPES IMPROVED FOR RESISTANCE/TOLERANCE TRAITS TO BIOTIC AND ABIOTIC STRESSORS UNDER CLIMATE CHANGE SCENARIOS|
|Supervisori e coordinatori interni:||BASSI, DANIELE|
|Data di pubblicazione:||17-mar-2017|
|Parole Chiave:||Ideotyping; global sensitivity analysis; Morris method; parameter distribution; Sobol’ method; WARM rice model; WOFOST-GT2; modelling; blast; Magnaporthe oryzae B. Couch; fungal pathogens; canopy structure; chalkiness; head rice; photosynthetic efficiency; genotype × environment interactions; spikelet sterility; cold tolerance; heat stress; Oryza sativa L.; rice; salinity; salt stress; breeding; climate change; crop adaptation; resistance; tolerance; modelling platform.|
|Settore Scientifico Disciplinare:||Settore AGR/02 - Agronomia e Coltivazioni Erbacee|
|Citazione:||IN SILICO IDEOTYPING: DEFINITION AND EVALUATION OF RICE IDEOTYPES IMPROVED FOR RESISTANCE/TOLERANCE TRAITS TO BIOTIC AND ABIOTIC STRESSORS UNDER CLIMATE CHANGE SCENARIOS / L. Paleari ; supervisor: R. Confalonieri ; coordinatore scuola di Dottorato: D. Bassi. - Milano : Università degli studi di Milano. DIPARTIMENTO DI ECONOMIA, MANAGEMENT E METODI QUANTITATIVI, 2017 Mar 17. ((29. ciclo, Anno Accademico 2016.|
|Digital Object Identifier (DOI):||http://dx.doi.org/10.13130/paleari-livia_phd2017-03-17|
|Appare nelle tipologie:||Tesi di dottorato|