Development and optimization of RNA switches for transcriptional control applying machine learning

• Grant number: 23/02133-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: October 01, 2023
• Status: Discontinued
• Field of knowledge: Biological Sciences - Genetics - Molecular Genetics and Genetics of Microorganisms
• Principal Investigator: Danielle Biscaro Pedrolli
• Grantee: Guilherme Engelberto Kundlatsch
• Host Institution: Faculdade de Ciências Farmacêuticas (FCFAR). Universidade Estadual Paulista (UNESP). Campus de Araraquara. Araraquara , SP, Brazil
• Associated scholarship(s): 25/02907-0 - Development of a high-throughput protocol for testing an RNA Switch Library for training a Machine Learning Model, BE.EP.DR

Abstract

Toehold switches are synthetic elements designed to detect the presence of a trigger RNA and control gene expression at the translational level. These switches have found diverse applications, including serving as biosensors, enzyme screening tools, and regulatory elements for metabolic engineering. Only recently were toehold switches capable of performing transcriptional control designed. These are still in the early stages of development and present low sensitivity to the presence of triggers. The objective of this study is to optimize this emerging class of switches through the use of machine learning. To achieve this, a simplified protocol will be developed to evaluate in vitro transcription using the fluorescent RNA corn, which will be utilized to assess a library of 150 to 200 distinct switches. These switches will be designed using the NUPACK software and will possess varying CG content and free energy of formation. The fluorescence quantification obtained after the transcription of different switches will be employed to train a machine learning algorithm, which will subsequently be utilized to design novel, optimized switches. The performance of these novel switches will be analyzed both in vitro and in vivo. In addition to contributing to the development of transcriptional toehold switches, this study seeks to create software that can be employed by other researchers in the design of RNA switches. (AU)