Bayesian methods for analysis of stock mixtures from genetic characters


Pella, Jerome, and Michele Masuda
Cover date: 

An implementation of Bayesian methods to assess general stock mixtures is described. An informative prior for genetic characters of the separate stocks in a mixture is derived from baseline samples. A neutral, low-information prior is used for the stock proportions in the mixture. A Gibbs sampler—the data augmentation algorithm—is used to alternately generate samples from the posterior distribution for the genetic parameters of the separate stocks and for the stock proportions in the mixture. The posterior distribution incorporates the information about genetic characters in the baseline samples, including relatedness of stocks, with that in the stock-mixture sample to better estimate genotypic composition of the separate stocks. Advantages over usual likelihood methods include greater realism in model assumptions, better flexibility in applications, especially those with missing data, and consequent improved estimation of stock-mixture proportions from the contributing stocks. Two challenging applications illustrate the technique and its advantages.