A new approach for HIV-1 protease cleavage site prediction combined with feature selectionReportar como inadecuado




A new approach for HIV-1 protease cleavage site prediction combined with feature selection - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Acquiredimmunodeficiency syndrome AIDS is a fatal disease which highly threatens thehealth of human being. Human immunodeficiency virus HIV is the pathogeny forthis disease. Investigating HIV-1 protease cleavage sites can help researchersfind or develop protease inhibitors which can restrain the replication of HIV-1,thus resisting AIDS. Feature selection is a new approach for solving the HIV-1protease cleavage site prediction task and it’s a key point in our research.Comparing with the previous work, there are several advantages in our work.First, a filter method is used to eliminate the redundant features. Second,besides traditional orthogonal encoding OE, two kinds of newly proposedfeatures extracted by conducting principal component analysis PCA andnon-linear Fisher transformation NLF on AAindex database are used. The twonew features are proven to perform betterthan OE. Third, the data set used here is largely expanded to 1922 samples.Also to improve prediction performance, we conduct parameter optimization forSVM, thus the classifier can obtain better prediction capability. We also fusethe three kinds of features to make sure comprehensive feature representationand improve prediction performance. To effectively evaluate the predictionperformance of our method, five parameters, which are much more than previouswork, are used to conduct complete comparison. The experimental results of ourmethod show that our method gain better performance than the state of artmethod. This means that the feature selection combined with feature fusion andclassifier parameter optimization can effectively improve HIV-1 cleavage siteprediction. Moreover, our work can provide useful help for HIV-1 protease inhibitordeveloping in the future.

 

KEYWORDS

Dimensionality Reduction; Machine Learning; HIV-1 Protease; Feature Fusion

Cite this paper

Yuan, Y. , Liu, H. and Qiu, G. 2013 A new approach for HIV-1 protease cleavage site prediction combined with feature selection. Journal of Biomedical Science and Engineering, 6, 1155-1160. doi: 10.4236-jbise.2013.612144.





Autor: Yao Yuan, Hui Liu, Guangtao Qiu

Fuente: http://www.scirp.org/



DESCARGAR PDF




Documentos relacionados