Understanding plant stress responses: using systems biology approach and text mining methods
Date
2022
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The world’s population is growing exponentially, with a current growth rate of approximately 1.1% per year. As of 2017, the number of undernourished people in the world was estimated as 821 million (FAO). Climate variability is increasingly viewed as a significant cause of hunger. Due to climate change and global warming, different biotic and abiotic stresses pose a severe threat to the agricultural sector limiting crop productivity worldwide. In the natural environment, plants face multiple biotic and abiotic stresses and the combined effect of these stresses has a tremendous impact on crop yield. In this regard, it is important to take steps for a genome-scale molecular understanding of stress response mechanisms in plants to help develop stress-tolerant cultivars. The amount of scientific literature on plant stress responses keeps increasing and this could pose a challenge to researchers as important information could be buried in the text. Biologists need to obtain a comprehensive knowledge of biological systems. For this reason, an approach to combine our knowledge in ‘omics’ studies and text mining to link genes to their function in plants when imposed with environmental stress has been implemented. The overarching objective of this dissertation is to improve our understanding of stress response in plants using ‘omics’ technologies and to complement standard enrichment analysis with text mining methods ☐ • First, RNA-Seq approach was used to understand the molecular mechanisms underlying stress response in an important bioenergy crop switchgrass (Panicum virgatum L.). Switchgrass was exposed to a single drought (DT) treatment and combinations of DT and heat (HT) (DTHT) stress treatment at different times points. Unique and overlapping genes and pathways were identified in response to DT and combined DTHT stress. ☐ • Secondly, we established a pipeline to automatically retrieve information on plant stress from the scientific literature to support the annotation of switchgrass. This pipeline integrates data from relevant resources to efficiently retrieve publications to study stress response in plants. The data collected is stored in MongoDB and used to predict additional role of the stress-responsive genes in switchgrass from the first study. We validated a candidate gene, Phenylalanine ammonialyase 1, involved in stress response in switchgrass. A preliminary work was conducted by evaluating in-house and publicly available tools to build a pipeline to retrieve literature to study stress response in the model plant Arabidopsis. ☐ • Lastly, to support the enrichment analysis performed in the first study, we created and visualized a functionally organized group of terms and pathways using ClueGO. The differentially expressed genes (DEGs) of the switchgrass transcriptome data was uploaded into ClueGO, a plugin of Cytoscape software. ClueGO integrates files from Gene Ontology, KEGG and Reactome, they were used to perform a ClueGO network of terms and pathways. ☐ The approach of combining systems biology and text mining methods to study stress response has generated valuable data to complement existing knowledge on plant stress. Such knowledge will eventually be useful to create a resource for the plant biology community and help with crop improvement in the long term.
Description
Keywords
Abiotic stress, Biotic stress, Switchgrass, Systems biology, Text mining, Transcriptomics