Comparison among Feature Encoding Techniques for HIV-1 Protease Cleavage Site Prediction

Kosesoy T., Gok M., Avci C.

9th International Conference on Electronics Computer and Computation (ICECCO 2012), Ankara, Türkiye, 1 - 03 Kasım 2012, ss.5-8, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Ankara
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.5-8
  • Kocaeli Üniversitesi Adresli: Evet

Özet

Recently, several works have approached the HIV-1 protease specificity problem by applying a number of methods from the field of machine learning. However, it is still difficult for researchers to choose the best method due to the lack of an effective comparison. In this paper, orthonormal encoding, N grams, composition moment vector, methods used for feature extraction from amino acid sequences. We have predicted HIV-1 protease cleavage sites by using SVM (with linear and radial basis kernel functions), IBK, BayesNet and K-star machine learning algorithms and the accuracies of classifiers have improved by utilization of Rotation Forest ensemble classification strategy. The results of the experiments evaluated by two metrics; classification accuracy and area under the receiver operating characteristic curve.