ABSTRACT
This chapter describes the likelihood approach, the nonlinear regression approach and the linear regression approach based on the expected cosegregation models among the two flanking markers and the quantitative trait loci (QTL). According to methods used for parameter estimation, interval mapping methods can be classified into three categories: likelihood approach, regression approach, and a combination of the likelihood and the regression approaches. The likelihood function for interval mapping is based on cosegregation among the putative QTL and the two flanking markers. The nonlinear regression model can be easily extended to multiple environments data. The chapter discusses a likelihood approach for QTL mapping in an F2 progeny set using codominant markers and then using dominant markers. Composite interval mapping (CIM) is a combination of simple interval mapping and multiple linear regression. The CIM can be implemented using the linear regression model for interval mapping and the multiple linear model to control for the residual genetic effects.
