The purpose of this project is to develop a new computational algorithm and a new software package for genome-wide association studies (GWAS) for traits with a distribution other than normal. The model is called the generalized linear mixed model (GLMM). Traits that do not follow a normal distribution are very common in crops and farm animals. For example, disease resistance traits are typical traits targeted by GLMM.
The idea behind the GLMM is the proposed pseudo response variable that is assumed to follow a normal distribution and is fitted to the usual linear mixed model (LMM). This pseudo response variable is a function of the current parameters. Once the parameters are estimated, the pseudo response variable will be fitted again to update the parameters. The updated parameters are then used to update the pseudo response variable. The process continues until both the parameters and the pseudo response variable converge.