A DISCRETE APPROACH TO LINKAGE MAPPING: USER DEFINED MARKER RESOLUTION AND HONEST MARKER PLACEMENT

¹ USDA-ARS Center for Bioinformatics and Comparative Genomics Cornell University Ithaca NY 14853

² Dept. of Computer Science Cornell University Ithaca NY 14853

³ Dept. of Operations Research and Industrial Engineering Cornell University Ithaca NY 14853

⁴ Dept. of Plant Breeding Cornell University Ithaca NY 14853

The resolution of a linkage map may best be described by the distribution of distances between recombinational or radiation-induced breakpoints in the corresponding mapping population. In a whole-genome linkage map, it is possible to catalog the vast majority of the breakpoints in a population and to locate them with reasonable precision. This allows one to consider the problem of linkage mapping in terms of discrete bins, or stretches of genome bordered by breakpoints in at least one individual. Advantages of this conceptualization are several. For instance, linkage mappers have historically relied on large random samples of individuals to permit the relative placement of closely linked markers. However, for many high-density genome-wide marker mapping projects, map resolution is not best served by random sampling. We have developed heuristic methods that use a catalog of breakpoints to quickly and consistently find near-optimal mapping samples from larger mapping populations. In addition, we have developed a discrete approach to marker and bin ordering that is accurate, extremely fast to implement, and allows one to more directly convey the underlying information present in a linkage map than do conventional approaches.