Computing That Serves

Microarray Data Mining - Dealing with the Challenges


Thursday, December 8, 2005 - 10:00am


Gregory Piatetsky-Shapiro, President of KDnuggets

DNA Microarrays have a potential to revolutionize molecular biology and medicine, leading to new, more effective, and ultimately personalized drugs and treatments.  To realize this potential we need to overcome a number of challenges for microarray data analysis, including lack of absolute ground truth and having too few samples compared to too many genes.

Researchers analyzing the same data frequently come up with different classifications and different sets of marker genes and it is hard to determine who is right without expensive lab tests.  To address these problems we propose several approaches.  One is developing and using best practices methodology for data analysis.  Second is integrating additional data, including medical literature and gene onthology. Third is looking at gene groups, rather than at individual genes. Finally, we propose the development of a comprehensive synthetic microarray test suite, which would allow calibration of different methods on different types of data and determining which methods give the best results.


Gregory Piatetsky-Shapiro, Ph.D., ( is the President of KDnuggets, which provides research and consulting services in the areas of data mining, knowledge discovery, bioinformatics, and business analytics.  Gregory is also the Editor and Publisher of KDnuggetsTM News, the leading newsletter on data mining and knowledge discovery, and website, a top-ranked site for data mining and knowledge discovery, including software, jobs, courses, companies, publications, and more.

Gregory is the founder of Knowledge Discovery in Database (KDD) conference series – he organized and chaired the first three KDD workshops and later helped convert them into leading international conferences on data mining. He was elected in 2005 as the chair of ACM SIGKDD, the leading professional organization for Knowledge Discovery and Data Mining.

Gregory has over 60 publications, including 2 best-selling books and several edited collections on topics related to data mining and knowledge discovery.  He frequently gives invited talks on topics related to Data Mining and Knowledge Discovery.