An Image-Based Visualization of Microarray Features and Classification Results.

Peter Bajcsy and Lei Liu

Proceedings of the 10^th Intelligent Systems for Molecular Biology (poster), Edmonton, Alberta, Canada, 3-7 August 2002.

In this work, we present a novel visualization approach for fast screening and inspection of DNA microarray data. The novelty lies in viewing the DNA microarray data as high-dimensional images, including laser scanned imagery, extracted features and labeled classifi cation results. From a data analysis viewpoint, this type of display is very suitable for visual screening of errors and inspection of analyzed data because an image format is maintained throughout the entire analysis process. The microarray dots form a grid pattern, creating intrinsically grid-based information. Thus, it is natural to extract features and maintain the grid. Each extracted feature forms a point in the feature image which is then used for classifi cation and visual inspection.