About This Theme
Understanding and exploiting the interactions between crop genotypes, phenotypes, and microbial associations requires substantial investment in the development of a wide variety of novel computational methods, softwares, and systems.
The computational informatics theme provides the P2IRC research team with the infrastructure, tools, and techniques needed for high-throughput computing, analyzing and processing image data, exploring genotype-phenotype relationships, and collaborating.
Theme 3 will deliver the key to integrating phenotypic imagery with genomic information and it will ultimately allow non-specialists to use comprehensive phenotypic databases. By providing the computational services to enhance themes 1, 2, and 4 and foster program-wide collaboration, the impact on global food security will be greater than the sum of the parts.
P2IRC Cloud: Big Data Analytics for Crop Phenomics
This research provides the solution for P2IRC's need to efficiently and effectively collect, integrate, represent, store, process, and analyze large-scale crop phenomic data.
The P2IRC Cloud is a high-throughput scalable cloud framework for supporting crop phenomics research. P2IRC Cloud will integrate genotype, phenotype, and microbial information with environmental and agronomic information.
Researchers will be able to use cloud services for creating high speed processing pipelines to model, analyze, visualize, and explore multi-dimensional, temporal crop phenomic data in support of the P2IRC goal of revolutionizing crop breeding.
Data Analysis for Rapid Plant Phenotyping
This research will create new ways to automatically analyze images of plants and crops to identify traits related to growth, health, resilience, and yield.
Computer recognition has the potential to increase the speed, reliability, and precision of trait identification, which is currently done by people in the field. This automation will provide new opportunities in plant science and breeding to directly compare large numbers of individual crop plants with known genetic differences.
This project encompasses research across imaging technologies and image analyses and focuses on automatic trait recognition in field-grown, as opposed to laboratory-grown, crops to better capture how new breeds will grow in producers' fields. The ability to rapidly and accurately measure plant and crop traits from images is central to the P2IRC program and will play a pivotal role in the development of new computing tools to support the global community of plant breeders.
Genotype & Environment to Phenotype (GE2P)
GE2P focuses on computational approaches to decipher the relationships between genotype, environment, and phenotype. This research will develop P2IRC's foundational ontology for phenotypic traits.
In the first phase of the project, the associations between genotypic traits and phenotypes will be explored and several systems for predicting these associations will be investigated in parallel. These will include current approaches, such as techniques for GWAS (genome-wide association studies), and more novel approaches, such as deep learning and probabilistic graphical models.
GE2P will computationally link genotype to phenotype in a way that generates the best possible actionable insights to breeders to further their breeding programs.Project Leads
Systems and Collaboration
Current cyberinfrastructure tools are unable to provide the support needed to functionally link global communities of breeders and genomics researchers working with large datasets and novel analyses. This research will develop the applications and software infrastructure needed to bridge this gap and bring P2IRC technologies to project stakeholders and the global plant breeding community.
Project 3.4 will work to enhance access to evolving software and data for P2IRC researchers and manage the enormous digital assets of the research centre. Laptop, smartphone, and tablet-ready middleware will be built to ensure that data and services developed by P2IRC are ready for mobile app development. Infrastructure, interface techniques, and applications to support flexible search and visualization workflows for genetic and phenomic information will be created. To meet the needs of plant breeders in the field, the current practices of breeders and genetic scientists will inform the development of software and applications to support those tasks and collaborative workflows.
This project will deploy mobile and web-based applications for image search and exploration, genomic and phenotypic visualization, and broad collaboration across disciplines.
Mechanistic Modeling of Plant Development for Plant Phenomics
The objective of this project is to better understand the fundamental biological processes that control phenotypes and phenotypic variation of plants using data-driven mechanistic computational models of plant development.
This research will work to advance modeling methods including simulation programs, tools for model exploration and analyses, and model management software. Model creation will be enhanced by advancements in the understanding of select developmental processes in plants, such as the regulation of organ position and orientation and stress-induced architectural effects. Techniques for inferring the genotypes associated with specific phenotypes will be explored and established, based on the use of mechanistic models or 'reverse phenomics'.
This research will uncover new fundamental knowledge of plant development and has the potential to change the nature of plant biology from a descriptive science to a quantitative description of the relationship between genotype and phenotype. Project 3.5 will advance the application of mechanistic models to plant breeding.