A Bayesian approach to estimating the regression coefficients of a multinomial logit model

A Bayesian approach to estimation of the regression coefficients of a multinominal logit model with ordinal scale response categories is presented. A Monte Carlo method is used to construct the posterior distribution of the link function. The link function is treated as an arbitrary scalar function. Then the Gauss-Markov theorem is used to determine a function of the link which produces a random vector of coefficients. The posterior distribution of the random vector of coefficients is used to estimate the regression coefficients. The method described is referred to as a Bayesian generalized least square (BGLS) analysis. Two cases involving multinominal logit models are described. Case I involves a cumulative logit model and Case II involves a proportional-odds model. All inferences about the coefficients for both cases are described in terms of the posterior distribution of the regression coefficients. The results from the BGLS method are compared to maximum likelihood estimates of the regression coefficients. The BGLS method avoids the nonlinear problems encountered when estimating the regression coefficients of a generalized linear model. The method is not complex or computationally intensive. The BGLS method offers several advantages over Bayesian approaches. ^

Regulation of T cell function by complement C5A in mice during mycobacterial infection

Tuberculosis is the leading cause of death in the world due to a single infectious agent, making it critical to investigate all aspects of the immune response mounted against the causative agent, Mycobacterium tuberculosis , in order to better treat and prevent disease. Previous observations show a disparity in the ability to control mycobacterial growth between mouse strains sufficient in C5, such as C57BL/6 and B10.D2/nSnJ, and those naturally deficient in C5, such as A/J and B10.D2/nSnJ, with C5 deficient mice being more susceptible. It has been shown that during M. tuberculosis infection, C5 deficient macrophages have a defect in production of interleukin (IL)-12, a cytokine involved in the cyclical activation between infected macrophages and effector T cells. T cells stimulated by IL-12 produce interferon (IFN)-γ, the signature cytokine of T helper type 1 (Th1) cells. It is known that a cell-mediated Th1 response is crucial for control of M. tuberculosis in the lungs of humans and mice. This study demonstrates that murine T cells express detectable levels of CD88, a receptor for C5a (C5aR), following antigen presentation by macrophages infected with mycobacteria. T cells from C5 deficient mice infected with M. tuberculosis were found to secrete less IFN-γ and had a reduced Th1 phenotype associated with fewer cells expressing the transcription factor, T-box expressed in T cells (T-bet). The altered Th1 phenotype in M. tuberculosis infected C5 deficient mice coincided with a rise in IL-4 and IL-10 secretion from Th2 cells and inducible regulatory T cells, respectively. It was found that the ineffective T cell response to mycobacteria in C5 deficient mice was due indirectly to a lack of C5a via poor priming by infected macrophages and possibly by a direct interaction between T cells and C5a peptide. Therefore, these studies show a link between the cells of the innate and adaptive arms of the immune system, macrophages and T cells respectively, that was mediated by C5a using a mouse model of M. tuberculosis infection. ^