African Journal of Biotechnology Vol. 6 (6), pp. 650-657, 19 March 2007   

ISSN 1684–5315 © 2007 Academic Journals        

Review

Localizing genes using linkage disequilibrium in plants: integrating lessons from the medical genetics

Joel W. Ochieng^1*, Anne W. T. Muigai² and George N. Ude³

¹Section of Genetics, College of Agriculture and Veterinary Sciences, University of Nairobi, P.O. Box 29053 Nairobi, 00625 Kenya.

²Institute for Biotechnology Research, Jomo Kenyatta University of Agriculture and Technology, P.O. Box 62000 Nairobi, 00200 Kenya.

³Department of Natural Sciences, Bowie State University, 14000 Jericho Park Road, Bowie, MD 20715, USA.

*Corresponding author.  E-mail: jochieng@uonbi.ac.ke . Tel: +61 2 6620 3961, Fax: +61 2 6622 2080.

Accepted 9 February, 2007

Finding genes controlling quantitative traits will aid molecular breeding for crops and livestock with superior yields, growth rates, and evolutionary potential. Such genes can be located using the candidate gene approach, genome wide scans, or by within family mapping. Linkage disequilibrium (LD) or association mapping, is a candidate gene approach that relies on detecting a statistical association between the desired quantitative trait and a molecular marker allele. This approach is emerging as a leading tool for precise estimation of QTL positions, because it offers several advantages over family-based mapping: LD mapping detects associations with greater resolution, the associations detected are relevant population wide, and in plants, the use of natural populations would circumvent the need to raise large controlled crosses. However, LD approach is facing obstacles, with well over 60% of studies reporting associations in the medical genetics disapproved in subsequent tests. A large proportion of these false associations (or lack of it) result from population stratification, while the rest may be caused by other demographic and evolutionary processes that create a statistical association between a marker allele and the trait, such as bottlenecks, natural selection, hybridization and genetic drift. The problem is expected to escalate in plants, owing to the complex population structures. Regardless of the many recent methods that purport to take into account population stratification during association tests, we discuss the reasons why in plants, a priori knowledge of population structures is essential in any robust association analysis.

Key words: Association mapping, linkage disequilibrium, population structure, nonreplication, quantitative trait loci.