Navegando por Autor "Marques, Yuri Bento"

Agora exibindo 1 - 3 de 3

Erratum to: Mirnacle: machine learning with SMOTE and random forest for improving selectivity in pre-miRNA ab initio prediction
(BMC Bioinformatics, 2017) Marques, Yuri Bento; Oliveira, Alcione de Paiva; Vasconcelos, Ana Tereza Ribeiro; Cerqueira, Fabio Ribeiro
MicroRNAs (miRNAs) are key gene expression regulators in plants and animals. Therefore, miRNAs are involved in several biological processes, making the study of these molecules one of the most relevant topics of molecular biology nowadays. However, characterizing miRNAs in vivo is still a complex task. As a consequence, in silico methods have been developed to predict miRNA loci. A common ab initio strategy to find miRNAs in genomic data is to search for sequences that can fold into the typical hairpin structure of miRNA precursors (pre-miRNAs). The current ab initio approaches, however, have selectivity issues, i.e., a high number of false positives is reported, which can lead to laborious and costly attempts to provide biological validation. This study presents an extension of the ab initio method miRNAFold, with the aim of improving selectivity through machine learning techniques, namely, random forest combined with the SMOTE procedure that copes with imbalance datasets. By comparing our method, termed Mirnacle, with other important approaches in the literature, we demonstrate that Mirnacle substantially improves selectivity without compromising sensitivity. For the three datasets used in our experiments, our method achieved at least 97% of sensitivity and could deliver a two-fold, 20-fold, and 6-fold increase in selectivity, respectively, compared with the best results of current computational tools. The extension of miRNAFold by the introduction of machine learning techniques, significantly increases selectivity in pre-miRNA ab initio prediction, which optimally contributes to advanced studies on miRNAs, as the need of biological validations is diminished. Hopefully, new research, such as studies of severe diseases caused by miRNA malfunction, will benefit from the proposed computational tool.
Mirnacle: aprendizagem de máquina utilizando SMOTE e Random Forest para prover aumento da seletividade na predição ab initio de pre-miRNAs
(Universidade Federal de Viçosa, 2015-12-08) Marques, Yuri Bento; Cerqueira, Fábio Ribeiro; http://lattes.cnpq.br/1307057738481189
Os microRNAs (miRNAs) são importantes reguladores da expressão gênica em plantas e animais. Assim, miRNAs estão envolvidos na maioria dos processos biológicos, tor- nando o estudo dessas moléculas um dos temas mais relevantes da biologia molecular atualmente. Uma estratégia para encontrar novos miRNAs é procurar seus precursores (pre-miRNAs), que são estruturas ligeiramente maiores (70-120 nt) e têm uma estru- tura secundária na forma de hairpin (grampo de cabelo). No entanto, caracterizar pre-miRNAs in vivo ainda é uma tarefa complexa. Como consequência disto, méto- dos in silico foram desenvolvidos para prever a localização genômica de pre-miRNAs. No entanto, as ferramentas computacionais atuais têm problemas de seletividade, isto é, uma grande quantidade de falsos positivos é reportada. Este trabalho apresenta uma extensão do método desenvolvido por Tempel e Tahi, 2012, com o objetivo de melhorar a seletividade através da técnica de aprendizagem de máquina denominada Random Forest, combinada com o método SMOTE, que lida com conjuntos de dados desbalanceados. Comparando o método proposto com outras importantes abordagens na literatura, mostramos que os procedimentos descritos neste trabalho puderam me- lhorar substancialmente a seletividade, sem comprometer a sensibilidade. Para três conjuntos de dados utilizados nos experimentos realizados, a abordagem proposta al- cançou pelo menos 97 % de sensibilidade e proporcionou um aumento de duas, vinte e seis vezes na seletividade, respectivamente, em comparação com os resultados de ferramentas computacionais atuais.
Mirnacle: machine learning with SMOTE and random forest for improving selectivity in pre-miRNA ab initio prediction
(BMC Bioinformatics, 2016-12-15) Marques, Yuri Bento; Oliveira, Alcione de Paiva; Vasconcelos, Ana Tereza Ribeiro; Cerqueira, Fabio Ribeiro
MicroRNAs (miRNAs) are key gene expression regulators in plants and animals. Therefore, miRNAs are involved in several biological processes, making the study of these molecules one of the most relevant topics of molecular biology nowadays. However, characterizing miRNAs in vivo is still a complex task. As a consequence, in silico methods have been developed to predict miRNA loci. A common ab initio strategy to find miRNAs in genomic data is to search for sequences that can fold into the typical hairpin structure of miRNA precursors (pre-miRNAs). The current ab initio approaches, however, have selectivity issues, i.e., a high number of false positives is reported, which can lead to laborious and costly attempts to provide biological validation. This study presents an extension of the ab initio method miRNAFold, with the aim of improving selectivity through machine learning techniques, namely, random forest combined with the SMOTE procedure that copes with imbalance datasets. By comparing our method, termed Mirnacle, with other important approaches in the literature, we demonstrate that Mirnacle substantially improves selectivity without compromising sensitivity. For the three datasets used in our experiments, our method achieved at least 97% of sensitivity and could deliver a two-fold, 20-fold, and 6-fold increase in selectivity, respectively, compared with the best results of current computational tools. The extension of miRNAFold by the introduction of machine learning techniques, significantly increases selectivity in pre-miRNA ab initio prediction, which optimally contributes to advanced studies on miRNAs, as the need of biological validations is diminished. Hopefully, new research, such as studies of severe diseases caused by miRNA malfunction, will benefit from the proposed computational tool.