Outlier Detection in Microarray Data Using Hybrid Evolutionary Algorithm

Abstract

DNA microarray technology has emerged as a boon to the scientific community in understanding the growth and development of life as well as in widening their knowledge in exploring the genetic causes of anomalies occurring in the working of the human body. Microarray technology makes biologists be capable of monitoring expression of thousands of genes in a single experiment on a small chip. Extracting useful knowledge and information from these microarrays has attracted the attention of many biologists and computer scientists. Outlier is defined as an observation that deviates too much from other observations. The identification of outliers can lead to the discovery of useful and meaningful knowledge. In this paper we propose hybrid approach for outlier detection by studying the behavior of projections from the data set. We first identify patterns with abnormally low presence. Once such patterns have been identified, then the outliers are defined as those records which have such patterns present in them. We have compared time and quality aspects of our results and it is found to be better than both brute force and evolutionary algorithms. For the same data set, brute force algorithm takes more time with good quality results as compared to evolutionary algorithm, where as Evolutionary Algorithm takes less time with less quality compared to brute force algorithm.