A new era in plant functional genomics. Issue 15 (June 2019)

• Record Type: Journal Article
• Authors: Rai, Amit; Yamazaki, Mami; Saito, Kazuki
• Is Part Of: Current opinion in systems biology. Issue 15(2019)
• Page Start: 58
• Page End: 67
• Publication Date: 2019-06
• Subjects: Plant systems biology -- GWAS -- Integrative omics analysis -- Metabolic models -- Omics -- Legacy data for systems biology -- Machine learning for systems biology; Systems biology -- Periodicals; 570
• DOI: 10.1016/j.coisb.2019.03.005
• Languages: English
• ISSNs: 2452-3100

Abstract

Abstract: In the last few years, research in plant sciences has seen an unprecedented growth of high dimensional omics data sets and its application in functional genomics studies. These studies have used the individual or integrative omics analysis approach for establishing a network of relationships between biomolecules within a system. The expansion of knowledge base–derived and the generated omics data sets offer a unique opportunity to use deep learning approaches to derive plant metabolic models. The deep learning algorithms require large data sets to identify important associations that regulate a biological process and to achieve high accuracy against expected outcomes. Legacy omics data sets could serve as input for such deep learning algorithms, while knowledge base serves as its expected outcome around which the algorithm attempts to achieve high prediction accuracy. Therefore, structured metadata submission associated with omics data sets and knowledge derived from it is valuable to derive next-generation toolsets for functional genomics in plant sciences. Here, we have reviewed recent advances in the functional genomics knowledge base driven by results from genomics, transcriptomics, metabolomics analysis, and their integration. We have particularly focused on studies that have generated large-scale omics data sets that will be suitable for deriving the plant metabolic models. We conclude our review with a brief discussion on requirements for legacy data to derive metabolic models by using machine learning and deep learning approaches. Highlights: Comparative genomics to predict gene function. Expansion of transcriptome and metabolome resources. Functional genomics based on genome-wide association study and metabolome-centric genome-wide association study. Metabolic modeling and its reconstruction. Legacy omics data as constraint for metabolic model reconstruction.