2019/02/07 by Bryan D Martin, Daniela Witten, Martin, Bryan D. +3 · 3 citations
Computer Science · Mathematics · Medicine · #Bayesian Methods and Mixture Models #FOS: Computer and information sciences #Methodology (stat.ME) #Pneumonia and Respiratory Infections #Statistical Methods and Bayesian Inference
paper · pdf · doi:10.48550/arxiv.1902.02776
openalex publication_date 2019/02/07 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
Using a sample from a population to estimate the proportion of the population\nwith a certain category label is a broadly important problem. In the context of\nmicrobiome studies, this problem arises when researchers wish to use a sample\nfrom a population of microbes to estimate the population proportion of a\nparticular taxon, known as the taxon's relative abundance. In this paper, we\npropose a beta-binomial model for this task. Like existing models, our model\nallows for a taxon's relative abundance to be associated with covariates of\ninterest. However, unlike existing models, our proposal also allows for the\noverdispersion in the taxon's counts to be associated with covariates of\ninterest. We exploit this model in order to propose tests not only for\ndifferential relative abundance, but also for differential variability. The\nlatter is particularly valuable in light of speculation that dysbiosis, the\nperturbation from a normal microbiome that can occur in certain disease\nconditions, may manifest as a loss of stability, or increase in variability, of\nthe counts associated with each taxon. We demonstrate the performance of our\nproposed model using a simulation study and an application to soil microbial\ndata.\n