Efficient Analysis of Microarray Gene Expressions
A.K.M. Tauhidul Islam
Microarray Experiments Gene Association Analysis Gene Selection MapReduce;
- 원문 URL
Microarray data analysis has been widely used for extracting relevant biological information from thousands of genes simultaneously expressed in a specific cell. Although many genes are expressed in a sample tissue, most of these are irrelevant or insignificant for clinical diagnosis or disease classification because of missing values and noises. Thus, finding a small, closely related gene set for classification and extracting interesting gene associations are important research problems for microarray analysis. At the same time, scalable methods are required for efficient management of rapidly increasing large volume of microarray data. Moreover, the extracted knowledge should match with publicly available biological resources. In this dissertation, we have first proposed a scalable parallel gene selection method using MapReudce programming model. The proposed method utilizes kNN classifier algorithm for fitness classification. We have used several real and synthetic datasets for experiments. Experiment results show that the proposed method can offer good scalability on large data with increasing number of nodes and can also provide higher classification accuracy rather than using whole gene set. Then, we have explored high utility mining for Gene Association Analysis (GAA) to extract biologically interesting gene associations. We have introduced row enumeration based high utility mining to handle high dimension data. The generated rules are validated with publicly available domain knowledge to signify the effectiveness of the proposed GAA method.