Use of statistical tools and molecular networking for the metabolomics analysis of Colletotrichum spp

• Grant number: 18/03679-8
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Effective date (Start): July 05, 2018
• Effective date (End): July 04, 2019
• Field of knowledge: Physical Sciences and Mathematics - Chemistry - Organic Chemistry
• Principal Investigator: Roberto Gomes de Souza Berlinck
• Grantee: Mirelle Takaki
• Supervisor: Roger Linington
• Host Institution: Instituto de Química de São Carlos (IQSC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Research place: Simon Fraser University, Burnaby, Canada
• Associated to the scholarship: 15/14114-3 - Which metabolomics components influence the phytopathogenicity of Colletotrichum spp.?, BP.DR

Abstract

In recent years, liquid chromatography coupled to mass spectrometry has become an essential tool for metabolomics studies. Non-targeted metabolomics studies result in a large amount of data. Such mega data analyses make the evaluation and interpretation of the metabolomics outcomes highly complex. The correct annotation and subsequent unambiguous identification of metabolites from these data still represents one of the main challenges in this area. In an attempt to overcome these barriers, data pre-treatment methods, processing algorithms and bioinformatics tools have been developed. One of these tools is Molecular Networking (MN), which has been widely used in complex samples studies. Molecular Networking organizes MS/MS data into molecular networks, providing an overview of all compounds detected and the chemical relationships between them. In addition to accelerating the process of discovery and characterization of unknown molecules, MN offers new ways for the investigation of the metabolome of biological samples. Another approach is the Compound Activity Mapping platform that quickly identify and predict the biological function of new bioactive components in complex samples directly from primary screening data. This tool combines chemical and biological screening with high-resolution untargeted metabolomics analysis and allows users to select compounds based on biological and/or chemical novelty. Static and multivariate chemometrics methods are also widely used in metabolomic studies, since these methods allow the combination of biological information to the interpretation of the metabolomics data. This project aims at to perform metabolomics analysis of phytopathogenic and endophytic Colletotrichum strains, using UPLC-MS/MS data jointly with Compound Activity Mapping tool, Molecular Networking, statistical and chemometric analyses in order to identify which secondary metabolites produced by Colletotrichum fungi are responsible for their phytotoxic activities. (AU)