Kernel conditional quantile estimation via reduction revisited

Quantile regression refers to the process of estimating the quantiles of a conditional distribution and has many important applications within econometrics and data mining, among other domains. In this paper, we show how to estimate these conditional quantile functions within a Bayes risk minimization framework using a Gaussian process prior. The resulting non-parametric probabilistic model is easy to implement and allows non-crossing quantile functions to be enforced. Moreover, it can directly be used in combination with tools and extensions of standard Gaussian Processes such as principled hyperparameter estimation, sparsification, and quantile regression with input-dependent noise rates. No existing approach enjoys all of these desirable properties. Experiments on benchmark datasets show that our method is competitive with state-of-the-art approaches. © 2009 IEEE.

Caspase-3-dependent phagocyte death during systemic Salmonella enterica serovar Typhimurium infection of mice.

Growth of Salmonella enterica in mammalian tissues results from continuous spread of bacteria to new host cells. Our previous work indicated that infective S. enterica are liberated from host cells via stochastic necrotic burst independently of intracellular bacterial numbers. Here we report that liver phagocytes can undergo apoptotic caspase-3-mediated cell death in vivo, with apoptosis being a rare event, more prevalent in heavily infected cells. The density-dependent apoptotic cell death is likely to constitute an alternative mechanism of bacterial spread as part of a bet-hedging strategy, ensuring an ongoing protective intracellular environment in which some bacteria can grow and persist.