Most industrial Saccharomyces cerevisiae strains used in food or biotechnology processes are benign. However, reports of S. cerevisiae infections have emerged and novel strains continue to be developed. In order to develop recommendations for the human health risk assessment of S. cerevisiae strains, we conducted a literature review of current methods used to characterize their pathogenic potential and evaluated their relevance towards risk assessment. These studies revealed that expression of virulence traits in S. cerevisiae is complex and depends on many factors. Given the opportunistic nature of this organism, an approach using multiple lines of evidence is likely necessary for the reasonable prediction of the pathogenic potential of a particular strain. Risk assessment of S. cerevisiae strains would benefit from more research towards the comparison of virulent and non-virulent strains in order to better understand those genotypic and phenotypic traits most likely to be associated with pathogenicity.

In this paper, we develop an unsupervised generative clustering framework that combines variational information bottleneck and the Gaussian Mixture Model. Specifically, in our approach we use the variational information bottleneck method and model the latent space as a mixture of Gaussians. We derive a bound on the cost function of our model that generalizes the evidence lower bound (ELBO); and provide a variational inference type algorithm that allows to compute it. In the algorithm, the coders' mappings are parametrized using neural networks and the bound is approximated by Markov sampling and optimized with stochastic gradient descent. Numerical results on real datasets are provided to support the efficiency of our method.Preprint. Under review.

The rise of algorithmic decision making led to active researches on how to define and guarantee fairness, mostly focusing on one-shot decision making. In several important applications such as hiring, however, decisions are made in multiple stage with additional information at each stage. In such cases, fairness issues remain poorly understood. In this paper we study fairness in k-stage selection problems where additional features are observed at every stage. We first introduce two fairness notions, local (per stage) and global (final stage) fairness, that extend the classical fairness notions to the k-stage setting. We propose a simple model based on a probabilistic formulation and show that the locally and globally fair selections that maximize precision can be computed via a linear program. We then define the price of local fairness to measure the loss of precision induced by local constraints; and investigate theoretically and empirically this quantity. In particular, our experiments show that the price of local fairness is generally smaller when the sensitive attribute is observed at the first stage; but globally fair selections are more locally fair when the sensitive attribute is observed at the second stage-hence in both cases it is often possible to have a selection that has a small price of local fairness and is close to locally fair.