Harnessing multi-omics data to unravel the genetic and molecular basis of complex traits: A systems genetics study of maize drought response.
Meeting ID: 542 690 6103 Passcode: D0%QPXdN01
The defense will be held in French but the slides will be in English.
- Nicolas Langlade, Senior Researcher, INRAE - Reviewer
- Vincent Segura, Junior Researcher, INRAE - Reviewer
- Mikaël Lucas, Junior Researcher, IRD - Examiner
- Élisabeth Petit-Teixeira, Professor, University of Paris-Saclay - Examiner
- Andrea Rau, Senior Researcher, University of Paris-Saclay - Examiner
The Ph. D. project was supervised by
- Mélisande Blein-Nicolas, Research Engineer, INRAE, Thesis Director
- Marie-Laure Martin, Senior Researcher, INRAE, Thesis Co-director
system biology, genome-wide association study, multi-omics data integration, molecular networks, genotype-phenotype relationship, genotype-by-environment interaction
Although interactions between academic research and plant breeding companies over the last century have enabled significant progress in crop improvement by producing high-yielding varieties, it has been several years since major crop yields have increased significantly. In addition, according to the Intergovernmental Panel on Climate Change (IPCC), anthropogenic greenhouse gas emissions have triggered an irreversible rise in temperatures that will make agricultural land drier and significantly increase crop failures over the next 30 years. Feeding nearly ten billion people in the face of climate change is, therefore, one of the greatest challenges of this century. Among crops affected by drought, maize is at the center of research into improving varieties to make them more resilient to water stress. However, drought tolerance is a polygenic trait (i.e., a trait under the control of multiple genes) that is highly dependent on the environment, which makes understanding its genetic determinism a tremendous task. After perceiving water stress, plants trigger multiple molecular pathways that moderate root water uptake and leaf evapotranspiration, which can affect their development and reduce yield. Thanks to advances in biotechnology, it is now possible to generate multi-omics datasets that can be used to conduct systems genetic approaches to address the complexity of drought tolerance. My PhD thesis aimed to gain insight into the genetic and molecular basis of maize drought response by performing an integrative analysis of multi-omics data (genomics: one million SNPs, proteomics: 2,000 proteins, metabolomics: 1,500 metabolites, and phenomics: 6 drought-related ecophysiological traits) measured for 254 maize hybrids grown in a greenhouse under two contrasting water regimes. In the first part of the thesis, I focused on the analysis of phenomics data to quantify the relevance of integrating plasticity indices in genome-wide association studies to detect QTLs involved in the genotype-by-water availability interaction (GxW). The main result of this part was that plasticity QTLs do not overlap with QTLs detected on phenotypic means, and they exclusively capture an important part of the GxW variance (10-70% depending on the traits). Besides identifying novel genetic regions potentially involved in drought response, my results support the postulate that phenotypic plasticity is an independent trait with its own genetic determinisms. The latter could be advantageous in breeding in order to design high-yielding varieties that optimize their water management through plastic ecophysiological traits. In the second part of the thesis, I conducted a systems genetics approach by integrating genomics, proteomics, and phenomics to i) infer a multiscale network revealing the genetic and molecular basis of the response to water stress, ii) assess the contribution of proteomics data in explaining GxW variance, and iii) provide a functional annotation of QTLs maximizing the proportion of GxW variance captured. Firstly, I was able to identify genomic regions enriched in pQTLs and translate them into protein-protein interaction networks. This enabled me to show that proteins associated with pQTLs located in these regions could physically interact with proteins encoded by genes covered by these regions. Secondly, I identified a set of QTLs and pQTLs that together capture 84% of the GxW variance. Thirdly, I inferred a multi-scale network comprising 531 loci, 63 proteins, and 6 drought-responsive traits. These results highlight the potential of omics data to reveal the genetic and molecular basis of complex traits such as drought tolerance. Overall, the results of my thesis encourage the consideration of phenotypic plasticity and the use of omics data to facilitate the design of drought-tolerant varieties in molecular-assisted selection.