Dear Mclust users, Pardon me for naive question. Mclust defines clustering based on parameterized finite Gaussian mixture models. The diabetes example data in vignette vectors glucose,.insulin and sspg are not exactly appear to be following a Gaussian distribution. I like to use mclust to cluster my data with two different observations. In my data, there are two observations. One observation follow a std.normal with right extended tail and second observation follow a negative binomial distribution. Is it appropriate to use mclustBIC Mclust and mclustBootstrapLRT? and If it is not an appropriate method and package, which package could help for my data. Thanks Adrian [[alternative HTML version deleted]]