Genomic analysis of clinical isolates of Fusarium spp. and genetic determinants of azole resistance using machine learning methods

• Grant number: 22/00754-4
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: October 01, 2022
• End date: June 12, 2023
• Field of knowledge: Biological Sciences - Microbiology - Applied Microbiology
• Principal Investigator: Marcia Regina von Zeska Kress
• Grantee: Otávio Guilherme Gonçalves de Almeida
• Host Institution: Faculdade de Ciências Farmacêuticas de Ribeirão Preto (FCFRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil

Abstract

Fusarium spp. are opportunistic fungi that cause a spectrum of infections ranging from Onychomycosis and Keratitis to systemic infections that affect especially immunosuppressed patients. The identification of Fusarium species is laborious and complex, involving molecular and phenotypic schemes and, even so, it is not resolute due to the cryptic speciation of these microorganisms. In addition, these fungi are notoriously recognized for their intrinsic resistance to most of the antifungals used, especially azoles. Little is known about the molecular mechanisms that allow these microorganisms to tolerate and to resist these compounds, and it is postulated that the presence of cyp51 genes as well as other genetic determinants, such as transport proteins, are the effectors involved in this process. In this project, the objective is to characterize, from the perspective of machine learning, which genetic determinants (i.e. cyp51, transporters, etc.) are related to the different levels of resistance of clinical isolates of Fusarium spp. to azoles and to establish a pipeline that makes it possible, from complete genome data, to characterize Fusarium isolates according to their resistance to azoles. For this, publicly available Fusarium genomes containing associated metadata in the articles that gave rise to their publication will be selected to feed the machine learning model. In addition, eight genomes from the FCFRP-USP culture collection will be selected for whole-genome sequencing, in addition to contributing to machine learning analyses, to be publicly available in public repositories. At the same time, given the importance of cyp51 in the resistance process, in the new sequenced genomes, paralogous sequences will be investigated for the design of primers for new markers involving this gene. Finally, it is expected that new genetic determinants involved in Fusarium spp. resistance to azoles be identified and to make it possible to classify and group fungal genomes based on their azole resistance profile using machine learning methods. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DE ALMEIDA, OTAVIO GUILHERME GONCALVES; KRESS, MARCIA REGINA VON ZESKA. Harnessing Machine Learning to Uncover Hidden Patterns in Azole-Resistant CYP51/ERG11 Proteins. MICROORGANISMS, v. 12, n. 8, p. 17-pg., 2024-08-01. (22/00754-4, 20/07546-2, 23/12463-7)