Plant & Animal Genomes XIII Conference

January 15-19, 2005
Town & Country Convention Center
San Diego, CA

in silico Mapping From Phenotypic, Pedigree, And Genomic Data In Self-Pollinated Crops

Martin Arbelbide¹ , Jianming Yu² , Rex Bernardo¹

¹  Department of Agronomy and Plant Genetics, University of Minnesota, 1991 Upper Buford Cr., St. Paul, Minnesota 55108

²  Institute for Genomic Diversity, Cornell University, 157 Biotechnology Building, Ithaca, New York, 14853

Most agronomically important traits are quantitative: they are characterized by a continuum of phenotypes, controlled by many genes and strongly influenced by the environment. Quantitative trait loci (QTL) mapping relies on finding markers closely linked to genes of interest. Traditional QTL mapping studies have used designed plant populations obtained by crossing contrasting inbred lines. Small population sizes and limited phenotypic evaluation limits the power to detect QTL in these populations. Plant breeding programs generate massive amounts of information not currently used in QTL mapping analyses. This information consists of phenotypic data from performance trials, pedigree records, and molecular marker genotypes generated from genomic screens. The objective of this study is to develop computational methods to map QTL using information routinely generated in plant breeding programs. An in silico mapping approach via a mixed-model analysis will be applied to computer-simulated data of a breeding program for soybean, a major self-pollinated crop. The usefulness of the approach, in terms of the power to detect and estimate QTL effects, will be studied under different levels of heritability, number of QTL, number of markers and population sizes.