Multiple classifier integration for the prediction of protein structural classes

Abstract

Supervised classifiers, such as artificial neural network, partition trees, and support vector machines, are often used for the prediction and analysis of biological data. However, choosing an appropriate classifier is not straightforward because each classifier has its own strengths and weaknesses, and each biological dataset has its own characteristics. By integrating many classifiers together, people can avoid the dilemma of choosing an individual classifier out of many to achieve an optimized classification results (Rahman et al., Multiple Classifier Combination for Character Recognition: Revisiting the Majority Voting System and Its Variation, Springer, Berlin, 2002, 167-178). The classification algorithms come from Weka (Witten and Frank, Data Mining: Practical Machine Learning Tools and Techniques, Morgan Kaufmann, San Francisco, 2005) (a collection of software tools for machine learning algorithms). By integrating many predictors (classifiers) together through simple voting, the correct prediction (classification) rates are 65.21% and 65.63% for a basic training dataset and an independent test set, respectively. These results are better than any single machine learning algorithm collected in Weka when exactly the same data are used. Furthermore, we introduce an integration strategy which takes care of both classifier weightings and classifier redundancy. A feature selection strategy, called minimum redundancy maximum relevance (mRMR), is transferred into algorithm selection to deal with classifier redundancy in this research, and the weightings are based on the performance of each classifier. The best classification results are obtained when 11 algorithms are selected by mRMR method, and integrated together through majority votes with weightings. As a result, the prediction correct rates are 68.56% and 69.29% for the basic training dataset and the independent test dataset, respectively. The web-server is available at http://chemdata.shu.edu.cn/protein_st/.

Cancer Classification from Gene Expression Data by NPPC Ensemble

“…Additionally, multiclass cancer classification is possible in this frame work by extending the binary NPPC to multiclass NPPC. The proposed NPPC ensemble framework may well be of interest to others for noisy data sets in the fields of machine learning and computational biology, such as detection of horizontal gene transfer in bacterial genomes [88], classification of functional classes of proteins [89], classification of the nature of the infectious diseases [90], diagnosis of the genetic abnormalities [91], and other important areas of medical diagnostics.…”

Accurate prediction of protein structural class using auto covariance transformation of PSI-BLAST profiles

“…These algorithms include neural network (Cai and Zhou 2000), support vector machine (SVM) (Cai et al 2001;Chen et al 2006a;Li et al 2008;Qiu et al 2009), fuzzy k-nearest neighbor (Zhang et al 2008, fuzzy clustering (Shen et al 2005), Bayesian classification (Wang and Yuan 2000), logistic regression (Kurgan and Chen 2007;Kurgan and Homaeian 2006), rough sets (Cao et al 2006) and classifier fusion techniques (Cai et al 2006;Chen et al 2006bChen et al , 2009Feng et al 2005;Kedarisetti et al 2006). Among them, SVM is the most popular and the best-performing classifier for this task (Kurgan et al 2008a).…”

Prediction of protein structural class based on Linear Predictive Coding of PSI-BLAST profiles

“…For convenience, the PSSM is denoted as is to feed these features to an appropriate classification algorithm to efficiently and accurately predict structural class. Up to now, a lot of machine-learning algorithms have been proposed, such as neural network [21], support vector machine (SVM) [22][23][24][25], fuzzy clustering [26], fuzzy k-nearest neighbor [27,28], Bayesian classification [29], logistic regression [30], rough sets [31] and classifier fusion techniques [32][33][34][35][36]. Among the aforementioned classification algorithms, SVM is the most reliable and attained excellent performance on the SCOP problem [19].…”