Content‐Based Retrieval of Microarray Experiments

• Published in: Pattern Recognition in Computational Molecular Biology pp. 315 - 334
• Main Author: Oğul, Hasan
• Format: Book Chapter
• Language: English
• Published: Hoboken, NJ, USA John Wiley & Sons, Inc 19.11.2015
• Subjects: content‐based information retrieval; information retrieval performance evaluation; microarray experiments; software tools
• ISBN: 9781118893685; 1118893689
• DOI: 10.1002/9781119078845.ch17

This chapter introduces the basic terminology and techniques in conventional information retrieval systems. It gives an elaborate view of the concept of content‐based information retrieval. Information retrieval systems can be distinguished into Boolean retrieval and ranked retrieval systems. The chapter relates these concepts and methods of information retrieval to microarray gene expression analysis. It presents an introduction to the similarity metrics that would be particularly useful for comparing microarray experiments. Microarray technology enables measuring the expression levels of thousands of genes simultaneously. Content‐based retrieval of microarray experiments requires constructing fingerprints that represent the experiment content in a computational framework. The fingerprint refers analogously to the term of index in traditional information retrieval with some domain‐specific differences. The chapter outlines briefly the methods for evaluating information retrieval performance. Finally, the existing software tools are listed with their brief descriptions.