Plant, Animal & Microbe Genomes X Conference

January 12-16, 2002
Town & Country Convention Center
San Diego, CA

Workshop: Bioinformatics/Computers
            

DATA MINING AND ITS APPLICATION TO GERMPLASM ANALYSIS

¹ John Innes Centre, Norwich Research Park, Colney, Norwich, NR4 7UH, UK.

Data mining techniques have been used for knowledge discovery in many sectors including insurance, retail, meteorology and medicine to name just a few. These techniques exploit methodologies from several fields including statistics, artificial intelligence, information theory, uncertainty theory, fuzzy logic and heuristics. They are used for applications such as clustering data objects, classifying objects into groups or hierarchical structures, regression and association. Until recently, data mining techniques had not been used widely for biological research. Recent applications include clustering expression profiles in a microarray analysis. Statistical techniques are widely used to study linkage disequilibrium between markers and the linkage disequilibrium mapping of traits is an important area of research. A new European/US project, GENE-MINE (http://www.gene-mine.org/) is developing bioinformatics tools for the analysis of germplasm data. Within this project we are proposing that non-statistical data mining techniques can be used to study genome wide linkage disequilibrium and possibly map complex traits from haplotype data.