Machine learning allied to ultrasensitive plasmonic and nanopore sensors for neurochemical monitoring

• Grant number: 22/11819-0
• Support Opportunities: Scholarships abroad - Research
• Start date: April 01, 2023
• End date: January 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Chemistry - Analytical Chemistry
• Principal Investigator: Javier Erick Lobatón Villa
• Grantee: Javier Erick Lobatón Villa
• Host Investigator: Joshua Edel
• Host Institution: Instituto de Química (IQ). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Institution abroad: Imperial College London, England

Abstract

As the global number of cases of neurodegenerative diseases is constantly rising, a better understanding of the biochemical processes involved in the brain is an urgent task for early-stage diagnostics, prevention, and improvements in treatment efficiency. Dopamine is a key neurotransmitter that regulates several physiological functions in the brain and has been pointed out, among other important metabolites, as a potential biomarker for neurodegenerative diseases. In this research project, we propose the development of a novel bioanalytical method for monitoring dopamine using ultrasensitive nanopore/surface-enhanced Raman spectroscopic (SERS) sensors and support vector machine (SVM). To this end, homogeneous gold nanoparticles will be synthesized and modified with two capture agents, which will allow designing an antibodyless sandwich assay by promoting nanoparticle dimers formation with dopamine molecules in the gaps. Taking advantage from the controlled dimers formation, optical (SERS) and electrical detection methods will be evaluated and compared in terms of sensitivity, selectivity, and repeatability. Additionally, SVM will be implemented to improve and automatize the conventional time-consuming data analysis used in ultrasensitive nanopore platforms. The proposed method will be validated though the calculation of figures of merit in the analysis of biofluids (e.g., cerebrospinal fluid). It is expected that this machine learning-assisted method might help to move SERS/nanopore sensors from a niche detection strategy to a general analytical technique for neurological disease studies and human health care. (AU)