Navegando por Palavras-chave "Barreira Genética"

Agora exibindo 1 - 1 de 1
  • Nenhuma Miniatura disponível
    Item
    Somente Metadadados
    Algoritmo para Predição de Seleção de Resistência Aos Inibidores de Ns5a do Vírus da Hepatite C
    (Universidade Federal de São Paulo (UNIFESP), 2020-12-14) Almeida, Douglas De Andrade De [UNIFESP]; Janini, Luiz Mario Ramos [UNIFESP]; Universidade Federal de São Paulo
    Summary Objective: To develop an algorithm that, based on the genetic sequence of the HCV infecting virus, can estimate which are the best therapeutic treatments with the least probability of resistance selection for NS5A inhibitors. Method: A phased algorithm was created to select attributes relevant to the study and further development of a machine learning model. The attributes used in this algorithm are the population frequency of the resistance codons, the HCV codon usage and the genetic barrier between the patient's codons and the resistance codons. Results: It was possible to cross-check information from the patient's infectious virus, with information from the medical literature to structure a database with predictive variables and a response variable related to the presence or absence of drug resistance. The model was able to predict with an AUC> 0.99 which characteristics of the virus cause resistance in certain drugs. Conclusion: Codon Usage parameters, population prevalence of codons and genetic barrier, proved to be good predictors of resistance. However, the limitation of the data source implies the possibility of overfitting, which can only be discarded and / or corrected with further studies in the area using similar methodology.