SUMMER, a shiny utility for metabolomics and multiomics exploratory research
- PDF / 2,686,342 Bytes
- 7 Pages / 595.276 x 790.866 pts Page_size
- 85 Downloads / 154 Views
SHORT COMMUNICATION
SUMMER, a shiny utility for metabolomics and multiomics exploratory research Ling Huang1 · Antonio Currais2 · Maxim N. Shokhirev1 Received: 11 August 2020 / Accepted: 19 November 2020 © The Author(s) 2020
Abstract Introduction Cellular metabolites are generated by a complex network of biochemical reactions. This makes interpreting changes in metabolites exceptionally challenging. Objectives To develop a computational tool that integrates multiomics data at the level of reactions. Methods Changes in metabolic reactions are modeled with input from transcriptomics/proteomics measurements of enzymes and metabolomic measurements of metabolites. Results We developed SUMMER, which identified more relevant signals, key metabolic reactions, and relevant underlying biological pathways in a real-world case study. Conclusion SUMMER performs integrative analysis for data interpretation and exploration. SUMMER is freely accessible at http://summer.salk.edu and the code is available at https://bitbucket.org/salkigc/summer. Keywords Integrative analysis · Metabolomics · Multiomics · Bioinformatics software · Interactive user interface · Web server
1 Introduction With recent advances in untargeted metabolomics, thousands of steady-state metabolites can be profiled simultaneously in a standardized setting (Patti et al. 2012). After proper pre-processing and normalization, statistical analyses such as univariate/multivariate or regression analysis can identify significantly differentially expressed (DE) metabolites between a reference and a perturbed test condition (Cambiaghi et al. 2017). A wide range of tools have been developed to analyze the spectrum data, pre- and post-processing the data, and perform statistical analysis to find the DE metabolites (Forsberg et al. 2018; Spicer et al. 2017). Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11306-020-01750-7) contains supplementary material, which is available to authorized users. * Maxim N. Shokhirev [email protected] 1
Razavi Newman Integrative Genomics and Bioinformatics Core, The Salk Institute for Biological Studies, La Jolla, CA 92037, USA
Cellular Neurobiology Laboratory, The Salk Institute for Biological Studies, La Jolla, CA 92037, USA
2
However, functional interpretation of those DE metabolites within diverse biological contexts remains a significant challenge. Traditional pathway enrichment analysis assumes that most members of a given pathway are affected in response to a specific perturbation (Khatri et al. 2012). While this can work for well for global changes (Marco-Ramell et al. 2018), this assumption may not be useful for datasets with smaller changes or when only a part of the pathway is affected. In addition, traditional pathway enrichment analyses do not consider the interconnected nature of metabolites through metabolic reactions, thus they are not able to capture changes in a network occurring across multiple pathways or within a subnetwork of an annotated pathwa
Data Loading...
