I need to compare gene expression differences between multiple line pairs of alcohol preferring and non-preferring rat lines. I have 5 such line pairs, 3 are unrelated but two were derived independently from the same parent stock. For each line, there are 10 samples. I'll be testing multiple genes, but for simplicity assume just one gene whose expression is measures as geneExpression. Alcohol Preferring Alcohol Non-Preferring Line Pair X or Non-X Line 1a Line 1b 1 Non-X Line 2a Line 2b 2 Non-X Line 3a Line 3b 3 Non-X Line X4a Line X4b X4 X Line X5a Line X5b X5 X If all the line pairs were independently derived, a model could be geneExpression ~ Line.Pair + AlcoholPreference with the factor of interest being Alcohol Preference but, there is the "X" factor, with the 2 "X strain-pairs being related, whereas the others are unrelated to each other and also to the 2 "X" strain-pairs. We want to take into account the fact that there are really only 4 parent populations of these 5 strain-pairs so as to decrease the weighting put on the "X" strains in the model. What would the most appropriate approach to this be and how would the model be written? Thanks, Mark Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry Indiana University School of Medicine 15032 Hunter Court, Westfield, IN 46074 (317) 490-5129 Work, & Mobile & VoiceMail (317) 399-1219 Skype No Voicemail please [[alternative HTML version deleted]]