Integration of Computational Tools to Explore the Diversity of Temporal Regulation in Plant-Specialized Metabolism
Authors:
Samuel Seaver2* ([email protected]), Angela Ricono1, Sara El Alaoui2, Kathleen Greenham1 (PI)
Institutions:
1University of Minnesota; 2Argonne National Laboratory
Abstract
Plants produce an amazing diversity of specialized metabolites (SM) that offer many benefits to human society. SMs are essential for pharmaceutical products and non-medicinal applications in the chemical industry, food additives, dyes, perfumes, cosmetics, and nutraceuticals. These products offer the potential to increase the return on investment of current biofuel crops by providing high-value co-products. While many specialized metabolic enzymes have been characterized, their spatial and temporal regulation is less understood, creating a challenge for engineering and optimizing metabolite levels. Understanding how diverse plants differentially regulate the production of the products arising from the same SM pathway will enable researchers to engineer such plants with greater reliability. The goal of this project is to build a computational tool in DOE Systems Biology Knowledgebase (KBase) that would enable researchers to integrate transcriptome data with metabolic networks of general and specialized metabolism for different plant species. This tool will enable researchers to explore different combinations of SM precursors and identify key enzyme targets for engineering.
The research team aims to build a set of classifiers through the application of machine learning techniques that would enable this prediction. To do so, the team will focus on the glucosinolate (GSLs) class of SMs within the plant order Brassicales. This project’s objectives are to (1) experimentally design and benchmark the biosynthesis of multiple GSLs in eight phylogenetically distinct species from diverse families within the Brassicales using high resolution time series datasets; (2) reconstruct the general and specialized metabolic networks for GSL biosynthesis, enabling the integration of omics data; (3) train and test the model to predict GSL biosynthesis; and (4) use the KBase platform to encode this approach in a series of apps that will enable other researchers to apply this approach to their pathway of interest. To disseminate the utility of the tool for target identification for SM production, the team will host virtual and onsite training workshops. This will help to spur research into engineering plants as platforms for co-production of biofuel and co-products and also increase the plant user community on KBase.