Abstract:Micro array data provides information of expression levels of thousands of genes in a cell in a single experiment. Numerous efforts have been made to use gene expression profiles to improve precision of tumor classification. In our present study we have used the benchmark colon cancer data set for analysis. Feature selection is done using t -statistic. Comparative study of class prediction accuracy of 3 different classifiers viz., support vector machine (SVM), neural nets and logistic regression was performed using the top 10 genes ranked by the t-statistic. SVM turned out to be the best classifier for this dataset based on area under the receiver operating characteristic curve (AUC) and total accuracy. Logistic Regression ranks as the next best classifier followed by Multi Layer Perceptron (MLP). The top 10 genes selected by us for classification are all well documented for their variable expression in colon cancer. We conclude that SVM together with t-statistic based feature selection is an efficient and viable alternative to popular techniques.

Classification of large grass genome sequences has major challenges in functional genomes. The presence of motifs in grass genome chains can make the prediction of the functional behavior of grass genome possible. The correlation between grass genome properties and their motifs is not always obvious, since more than one motif may exist within a genome chain. Due to the complexity of this association most pattern classification algorithms are either vain or time consuming. Attempted to a reduction of high dimensional data that utilizes DAC technique is presented. Data are disjoining into equal multiple sets while preserving the original data distribution in each set. Then, multiple modules are created by using the data sets as independent training sets and classified into respective modules. Finally, the modules are combined to produce the final classification rules, containing all the previously extracted information. The methodology is tested using various grass genome data sets. Results indicate that the time efficiency of our algorithm is improved compared to other known data mining algorithms.