Copasetic analysis: Automated analysis of biological gene expression images

Abstract

In the past decade computational biology has come to the forefront of the public's perception with advancements in domain knowledge and a variety of analysis techniques. With the recent completion of projects like the human genome sequence, and the development of microarray chips it has become possible to simultaneously analyse expression levels for thousands of genes. Typically, a slide surface of less than 24 cm2, receptors for 30,000 genes can be printed, but currently the analysis process is a time consuming semi-autonomous step requiring human guidance. The paper proposes a framework, which facilitates automated processing of these images. This is supported by real world examples, which demonstrate the technique's capabilities along with results, which show a marked improvement over existing implementations.