Detalhes bibliográficos
| Ano de defesa: |
2024 |
| Autor(a) principal: |
Oviedo, Anna Karolline Rubim Rodrigues |
| Orientador(a): |
Gomes, Patricia |
| Banca de defesa: |
Campos, Andréia da Silva Fernandes,
Carvalho, José Antônio Mainardi de,
Rech, Virginia Cielo,
Oliveira, Jivago Schumacher de |
| Tipo de documento: |
Tese
|
| Tipo de acesso: |
Acesso embargado |
| Idioma: |
por |
| Instituição de defesa: |
Universidade Franciscana
|
| Programa de Pós-Graduação: |
Programa de Pós-Graduação em Nanociências
|
| Departamento: |
Biociências e Nanomateriais
|
| País: |
Brasil
|
| Palavras-chave em Português: |
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
http://www.tede.universidadefranciscana.edu.br:8080/handle/UFN-BDTD/1355
|
Resumo: |
Due to the COVID-19 pandemic, the number of infections caused by SARS-CoV-2, and antibiotic-resistant bacteria such as S. aureus, and K. Pneumoniae (classified as SKAPE microorganisms) has increased considerably. Given this scenario, the development of antiviral, and antioxidant agents is encouraged. Nanotechnology makes it possible to obtain nanomaterials (metallic nanoparticles such as Ti, Zn, Ag, and Cu) with antioxidant and antiviral activity, capable of inhibiting microorganisms of the ESKAPE class. Furthermore, these metallic nanoparticles such as titanium nanoparticles (TiO2-NPs), can be synthesized by natural sources, such as plant extracts (leaves of the Japanese grape, Hovenia dulcis) containing flavonoids responsible for the reduction, stabilization, and nucleation of metallic precursors. Dihydromyricetin (DHM) is a flavonoid with antioxidant, antitumor, and anticancer properties. In this context, the present study aims to synthesize and characterize titanium nanoparticles functionalized with DHM (DHM@TiO2-NPs) for application as an antioxidant agent, and antimicrobial activity against three bacterial strains (E. coli, and P. aeruginosa). Moreover, verify the interaction between the SARS-CoV-2 glycoprotein SPIKE and its Delta and Omicron variants, through molecular docking. At the same time, machine and deep learning algorithms were applied to correlate the administration of vaccines against COVID-19 (3, 1, and 2) and patient parameters (alcohol or cigarette consumption, frequency of physical activity, history of illness, BMI) with neuropsychiatric development after contracting COVID-19, using the Random Forest (RF), Xtreme Gradients Boosting Machine (XGB), Artificial Neural Network (ANN) and Recurrent Neural Network (RNN) algorithms. The in-silico studies showed that DHM showed greater spontaneity and interaction with the Delta (∆G = -8.9 kcal mol-1 ), and Omicron (∆G = -7.4 kcal mol-1 ) variants than the SPIKE glycoprotein (∆G = -5.7 kcal mol-1 ) pure. In addition to presenting greater similarity than the substance of propolis, and galangin. Regarding the experimental studies, the DHM@TiO2-NPs were characterized by X-ray diffraction (XRD), scanning electron microscopy with electron gun emission (FEG-SEM), N2 porosimetry, dynamic light scattering (DLS) for measurement of zeta potential (PZ), and hydrodynamic diameter, where it was verified that a highly pure DHM@TiO2-NPs nanocomposite obtained, since only the anatase phase (associated with TiO2-NPs), and quercetin (in relation to DHM) were identified. Additionally, it was observed that the DHM@TiO2-NPs had a mesoporous structure, surface area equal to 10 m2g-1, and pore volume 0.07 cm3 g-1, respectively. Fourier Transform Infrared Spectroscopy (FTIR) identified C=O, C=C groups associated with DHM and Ti-O, with TiO2-NPs. The hydrodynamic diameter, and zeta potential reported for the nanocomposite were 318 nm, and -18.70 mV indicating physical-chemical stability. The DHM@TiO2-NPs showed antioxidant activity (total phenols, and flavonoids equal to 10.65 and 18.57 mg g-1 , respectively) resulting in DPPH radical neutralization (0.44 µmol g-1 ). Furthermore, DHM@TiO2-NPs showed antimicrobial activity against E. coli, and P. aeruginosa (24 µg mL-1 ). Therefore, this study confirms the potential of DHM@TiO2-NPs as an antioxidant, and antimicrobial agent, which can be applied as food packaging, antibacterial agents (sanitizers) and water treatment. The algorithms XGB (Accuracy: 88.50% for training data and 87.89% for test data) and ANN (Accuracy: 89.32% for training data and 86.11% for test data) showed the better performances in machine and deep learning studies, with ANN being used for predictions, due to the complexity of the data. In this way, a neural network with 17 input variables, 2 hidden layers (with 10 neurons in the first and 8 in the second layer) and 1 neuron as a response variable was obtained. Using the deep learning model, it was found that the development of neuropsychiatric sequelae strongly depends on the patient's disease history, frequency of physical activity and the brand of the administered vaccine, with the 2 vaccines considered the safest among those investigated and with a lower tendency for the development of sequelae in patients who contracted COVID-19 once or twice. Therefore, it is possible to characterize machine learning algorithms as excellent prediction and correlation tools between complex data, being capable of reducing the number of clinical tests, as well as reducing the operational costs of research related to the adverse effects of COVID-19 vaccines. |
