Banca de QUALIFICAÇÃO: MARFRAN CLAUDINO DOMINGOS DOS SANTOS

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
STUDENT : MARFRAN CLAUDINO DOMINGOS DOS SANTOS
DATE: 10/02/2022
TIME: 10:00
LOCAL: Vídeoconferência
TITLE:

Chemometric techniques in the detection of arboviruses in clinical samples

KEY WORDS:

Dengue. Chikungunya. EEM matrices. Multiway classification. PARAFAC-QDA.

PAGES: 206
BIG AREA: Ciências Exatas e da Terra
AREA: Química
SUBÁREA: Química Analítica
SPECIALTY: Métodos Óticos de Análise
SUMMARY:

The pandemic established by COVID-19 showed the world the need for techniques capable of detecting viruses quickly and simply, in order to start treatment as soon as possible. Significant attempts are being made around the world to develop a tool that, with a simple analysis, is able to distinguish between different viruses. The objective of this work was to provide a theoretical framework on the application of chemometric techniques in the field of virology, and to evaluate the application of second-order spectroscopic techniques in the detection of arboviruses in clinical samples. In this thesis there are 3 articles: 2 review articles and 1 research article. The first review article addresses the application of Fourier Transform MidInfrared Spectroscopy with Attenuated Total Reflection (ATR-FTIR) in virus identification, bringing the theoretical foundations and various applications carried out over the last few years, where the technique was applied ATR-FTIR together with multivariate classification techniques. In the second review article, in addition to the ATR-FTIR technique, the techniques of near infrared spectroscopy (NIR), Raman and molecular fluorescence spectroscopy applied in the detection of various viruses were also addressed. Among the multivariate techniques covered in these review articles can be found: Principal Component Analysis (PCA), Successive Projection Algorithm (SPA), Genetic Algorithm (GA), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Analyzes of Parallel Factors (PARAFAC) and Discriminant Analysis based on the Partial Least Squares method (NPLSDA). In the research article, molecular fluorescence spectroscopy was applied together with the multi-way analysis techniques PARAFAC-LDA, PARAFAC-QDA and N-PLSDA, as a new sensitive, specific, fast, simple and capable of identifying variations. spectra between clinical samples with and without dengue and chikungunya viruses. For this, excitationemission matrices (EEM) of 26 samples for each class (healthy x dengue x chikungunya) were used, totaling 78 samples. The samples were divided into training and test sets and, after building the models, the results were analyzed based on precision, sensitivity, specificity and Fscore calculations. It was observed that the most efficient model in the classification work was the PARAFAC-QDA, hitting 100% of the samples in the test set. The loads captured in PARAFAC were provided and can be used in future studies to prove the relationship (or not) of spectral data with the presence of viruses. Finally, it was observed that EEM matrices have the potential to be used in the near future for screening or diagnosing arboviruses, having advantages when compared to more conventional techniques

BANKING MEMBERS:
Externo ao Programa - 1913849 - EDGAR PERIN MORAES
Interno - 1803692 - FABRICIO GAVA MENEZES
Presidente - 1714946 - KASSIO MICHELL GOMES DE LIMA