I think I am missing something here. You have written a standard linear mixed model which can be handled by any mixed model routine. The only way you could gain by using limma is if there is also some "k" and a linear model for each of the k responses. (Usually "k" is the gene id, but it could be any indicator of a high dimensional response.) If there is such a "k", I think that you could use the "block" option in limma to fit this model. --Naomi At 04:47 PM 3/7/2006, pingzhao Hu wrote: >Hi, >Say two groups A and B. Each group has 10 individuals. >Assume that we have two variables: Y and X. >Therefore, the hierarchical linear model will like: > >Yij=B0i+B1Xij+ERRORij, (i=A,B; j=1,2,...,10) > >We are interested in testing the significance of B1. >Although some R packages can do this job, we are interested in >using the shrinkage property in LIMMA or other R packages. > >Thank you for your help in advance! > >Pingzhao Hu >The Center for Applied Genomics >The Hospital for Sick Children >Toronto, Canada > >_______________________________________________ >Bioconductor mailing list >Bioconductor at stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor Naomi S. Altman 814-865-3791 (voice) Associate Professor Dept. of Statistics 814-863-7114 (fax) Penn State University 814-865-1348 (Statistics) University Park, PA 16802-2111