Find Significant Gene Information Based on Changing Points of Microarray Data

For transformations, a set of new basis is normally chosen for the data. The selection of the new basis determines the properties that will be held by the transformed data. For wavelet transform, a set of wavelet basis aims to detect the localized features contained in microarray data. In this research, we investigate the performance of wavelet features based on wavelet detail coefficients at third level in wavelet space, which characterize the changing points of microarray data based on high-order information. In order to find the significant gene information, we reconstruct wavelet details based on detail coefficients. A genetic algorithm is used to select the best features from reconstructed details in original data space, and corresponding gene information is detected based on selected features. Experiments are carried out on four datasets and experimental results show that good performance is achieved based on twofold cross-validation experiments.