Field design and the search for quantitative trait loci in plants Kent M. Eskridge Biometry Department University of Nebraska - Lincoln An efficient, well-planned field experiment and use of a precise and unbiased statistical analysis are critical in maximizing the chances of detecting marker-trait associations in plant QTL research. However, development of efficient field designs is complicated by small trait mean differences between marker classes, large amounts of uncontrollable environmental and field variation and the need for a large number of lines to have adequate power. Incomplete block designs can be used to minimize the effects of uncontrollable variation and to maximize the chances of detecting QTLs. Unfortunately, many QTL field experiments are not conducted as incomplete block designs, which likely reduces the ability to detect important QTLs. In addition, QTL researchers often use statistical methods for identifying and mapping QTLs that ignore important sources of variation such as block, environment and genotype x environment interaction resulting in inefficient and sometimes biased results. The objectives of this paper are to (1) illustrate how use of inefficient field designs and/or inadequate statistical methods reduces the chances for detecting QTL and biases results, (2) discuss some efficient field designs and (3) describe appropriate statistical approaches in the search for QTLs. The conclusion is that use of appropriate field designs and statistical analyses can considerably increase the chances for detecting important QTLs in plants. |
Graybill Conference |