Plant & Animal Genomes XIV Conference Plant & Animal Genomes XIV ConferenceJanuary 14-18, 2006
Town & Country Convention Center
San Diego, CA
Alexander Kozik , Richard Michelmore
Contemporary molecular marker techniques can generate data for thousands molecular markers to be mapped simultaneously. Construction and validation of high density genetic maps is a challenge and requires robust, high-throughput approaches. As part of the Compositae Genome Project, we developed a suite of Python scripts for quality control of genetic markers, grouping and inference of linear order of markers on linkage groups. These scripts can be used in conjunction with other mapping programs or can be used as a stand-alone package. The suite consists of three programs: MadMapper_RECBIT, MadMapper_XDELTA and CheckMatrix. MadMapper_RECBIT analyses raw marker scores for recombinant inbred lines. MadMapper_RECBIT generates pairwise distance scores for all markers, clusters based on pairwise distances, identifies genetic bins, assigns new markers to known linkage groups, validates allele calls, and assigns quality classes to each marker based on several criteria and cutoff values. MadMapper_XDELTA utilizes new algorithm, Minimum Entropy Approach and Best-Fit Extension, to infer linear order of markers. MadMapper_XDELTA analyzes two-dimensional matrices of all pairwise scores and finds best map that has minimal total sum of differences between adjacent cells (map with lowest entropy). This approach scales well and can accommodate large numbers of markers, unlike some commonly used mapping programs. CheckMatrix serves as a visualization tool to validate constructed genetic maps. CheckMatrix generates graphical genotypes and two-dimensional heat plots of pairwise scores. Visualization of regions with positive and negative linkage as well as of allele fraction per marker simplifies genetic map validation without applying statistical approaches. Scripts are freely available at http://cgpdb.ucdavis.edu/XLinkage/MadMapper/