Improving Objectivity and Scalability in Protein Crystallization: Integrating Image Analysis With Knowledge Discovery
Issue No. 06 - November/December (2001 vol. 16)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/5254.972075
<p>This article describes issues related to integrating image analysis techniques with knowledge discovery and case-based reasoning. Although the work applies to many problem domains, here we focus on analyzing and classifying outcomes of protein crystallization experiments in high-throughput structural genomics. We apply the fast Fourier transform to analyze image content to extract important features of the spectrum. We use a combination of these features to classify crystallization experiments' outcomes. Although humans can analyze images more flexibly, a computational approach makes the process scalable and more objective. We evaluate the classification process and present results on how we can combine automatically extracted features to discover important crystallographic knowledge.</p>
J. I. Glasgow et al., "Improving Objectivity and Scalability in Protein Crystallization: Integrating Image Analysis With Knowledge Discovery," in IEEE Intelligent Systems, vol. 16, no. , pp. 26-34, 2001.