Mine, Model and Make to Discover (3M2D): A platform for prospecting bioactive substances in (meta)genomic and metabolomic data from microorganisms, validation in expression platforms and artisanal fermentation

• Grant number: 25/03943-0
• Support Opportunities: Regular Research Grants
• Start date: August 01, 2025
• End date: July 31, 2028
• Field of knowledge: Biological Sciences - Microbiology - Applied Microbiology
• Principal Investigator: Ricardo Roberto da Silva
• Grantee: Ricardo Roberto da Silva
• Host Institution: Faculdade de Ciências Farmacêuticas de Ribeirão Preto (FCFRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil
• Associated researchers: Danielle Biscaro Pedrolli ; María Eugenia Guazzaroni ; Norberto Peporine Lopes

Abstract

Studies of microbial communities associated with the most diverse environments have become increasingly important for the sustainable exploration of biodiversity. Understanding the set of genomes of these communities, called metagenomes, is essential to find relationships between the coding potential of genomes and their respective ecological niches. The numerous evolutionary adaptations of microbial communities to the most diverse environments make public data mining a source of potential biotechnological innovations for various applications. However, metagenomic data processing poses a series of challenges for mining. For data to be processed efficiently, it is necessary to use a series of integrated applications, allowing execution in a high-performance computing environment. Interactions between members of microbial communities, as well as with their environments, are mediated by a wide range of metabolites. The main experimental approach used to study microbial metabolomes is liquid chromatography coupled to tandem mass spectrometry. This technique allows great flexibility in the separation and detection of polar metabolites of different chemical classes. With the expansion of public data repositories containing spectrometry experiments, several methods for correlating molecular and gene data have emerged. Despite advances in bioinformatics for functional annotation of gene sequences and associated metabolites, a significant portion of these sequences remains functionally uncharacterized. New predictive models for functional annotation have been using underlying structural and functional features that are shared and preserved in amino acid sequences. Although mining and predictive tools are an essential step in the prospecting of biosynthetic pathways, the expression of many genes has a complex regulation and therefore requires a set of regulatory components, such as promoters and terminators, that enable their appropriate expression. Once platforms for prospecting, prediction, and gene expression are established, the need to test the action of microorganisms and their genes under various experimental conditions becomes a major obstacle to new discoveries. The automation of microbial cultures allows precise control of growth conditions, allowing a wide range of studies. The Laboratory of Computational Chemobiology (http://ccbl.fcfrp.usp.br/) has been consolidating its position in the field of metabolomics and, more recently, in the field of genomics. In its new phase of research, the laboratory intends to integrate exploratory and predictive methods with a set of validation methods that allow the hypotheses raised by these studies to be tested and shorten the cycles for generating new biotechnological targets and refining computational methods. This phase of research has already begun and the first studies have already been submitted for review. These studies include genome sequencing, integration of metabolomics and genomics, automation of microbial cultures and monitoring of metabolism using metabolomics. The present project foresees the expansion of these approaches, with the incorporation of biosynthesis platforms of candidates obtained from the mining and prediction stages, as a key complement to the methods for controlling microbial cultures. (AU)