FF - Trabalhos de Conclusão de Curso
URI Permanente para esta coleção
Navegar
Navegando FF - Trabalhos de Conclusão de Curso por Autor "Andrade, Carolina Horta"
Agora exibindo 1 - 1 de 1
Resultados por página
Opções de Ordenação
Item Integração de abordagens computacionais para identificar novos inibidores da proteína NS5 do vírus Zika(Universidade Federal de Goiás, 2020-12-14) Ramos, Paulo Ricardo Pimenta da Silva; Andrade, Carolina Horta; Andrade, Carolina Horta; Borba, Joyce Villa Verde Bastos; Neves, Bruno JuniorThe Zika virus (ZIKV), which caused recent outbreaks and epidemics in 2007, 2014 and 2015, is transmitted to humans mainly through the bite of female Aedes aegypti mosquitoes. In addition, transmission through blood, intercourse and lactation has been described in the literature. ZIKV infection has several consequences, mainly neurological serious, so far, there are no antivirals to fight infection or vaccines to prevent infection. Non-structural protein 5 (NS5), composed by the methyltransferase (MTase) and RNA polymerase (RdRP) domains, plays an essential role in the synthesis and stability of viral RNA, in addition to inhibiting the host's immune system, being a promising target for development of new antiviral. In this work, we perform the integration of computational approaches as quantitative structure activity relationships (QSAR) based on machine learning methods, molecular docking and search for similarity to identify new inhibitors of the NS5 protein from ZIKV. Due to the availability of data in the literature and the high sequential similarity between the NS5 sites of the dengue virus (DENV) and ZIKV, we initially searched for inhibitors of DENV NS5 protein in the PubChem and ChEMBL databases, to guide the screening of new NS5 inhibitors from ZIKV. 145 DENV NS5 inhibitors have been found in literature. We performed the virtual screening of these compounds using ZIKV phenotypic QSAR models. We also carry out molecular docking studies in the binding sites of the MTase and RdRP domains of ZIKV NS5. A total of 32 compounds were prioritized at this stage. Subsequently, a similarity search as performed in the commercial eMolecules database and the QSAR and docking steps were performed with the similar compounds. This analysis resulted in 4,953 similar compounds that also passed through the QSAR and docking filters. After this step, 176 compounds were selected as promising inhibitors of ZIKV NS5. We also submitted These compounds to Bayesian machine learning models to predict ZIKV activity and cytotoxicity, resulting in 44 compounds predicted to be active and non-cytotoxic to mammalian cells. These compounds were then submitted to the nAPOLI software for analysis of ligand-protein interactions. This step allowed the final selection of 14 promising compounds that will be acquired and experimentally validated, through ZIKV NS5 enzymatic assays and celular assays, that will be performed with collaborators.