Desenvolvimento e otimização de estratégias analíticas visando o diagnóstico de COVID-19 em amostras de urina
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Universidade Federal de Juiz de Fora (UFJF)
Abstract
COVID-19 (Coronavirus Disease) is now a part of a group of diseases that represents
an important threat to public health, mainly in tropical countries and with a large percentage of
low-income populations, as is the case from Brazil. COVID-19 was the cause of the recent
global pandemic that maintained a state of emergency for approximately 3 years. The disease
is a viral infection caused by the new SARS-CoV-2 that shares symptomatic manifestations
with other diseases, such as the ones caused by Arboviroses and Influenza for example, despite
requiring different treatments. In this context, assertive laboratory diagnosis tests that are
accessible to the affected population is necessary. Therefore, the investigation of new analytical
strategies applicable to the diagnosis of COVID-19 is the proposal of this work. Optimizations
involving collection, preparation and storage of biological samples were initially studied,
followed by the development of a series of analytical methods using a multiplatform
instrumental approach in association with data science for the investigative study of metabolites
present in urine of patients affected by COVID-19 infection in comparison with healthy
individuals, generating diverse and comprehensive data sets. Segmented methods for analyzing
organic acids, amino acids and creatinine by capillary electrophoresis were optimized and
validated., as well as comprehensive approaches using liquid chromatography coupled to highresolution mass spectrometry and infrared spectroscopy using computational tools and
experimental design aimed at searching for biomarkers. Qualitative results obtained by
capillary electrophoresis with ultraviolet detection applied to numerous groups of urine samples
with an accuracy of 1.22% and by infrared spectroscopy were applied to build machine learning
prediction models for COVID-19 with an accuracy of 77.9% and 81.0%, respectively, useful
as analytical screening routines. Quantitative analysis of creatinine in urine by capillary
electrophoresis with UV detection using internal standard calibration with automated injection
protocol were validated, obtaining analytical precision of 2.43%. The use of capillary
electrophoresis associated with mass spectrometry was also evaluated. The applicability of the
methods to different samples and other analytical fronts are also discussed throughout the text.
Description
Keywords
Citation
Collections
Endorsement
Review
Supplemented By
Referenced By
Creative Commons license
Except where otherwise noted, this item's license is described as Acesso Aberto

