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. ^

Prevalence and risk factors for polyomavirus reactivation in solid-organ transplantation

Background. Polyomavirus reactivation is common in solid-organ transplant recipients who are given immunosuppressive medications as standard treatment of care. Previous studies have shown that polyomavirus infection can lead to allograft failure in as many as 45% of the affected patients. Hypothesis. Ubiquitous polyomaviruses when reactivated by post-transplant immunosuppressive medications may lead to impaired renal function and possibly lower survival prospects. Study Overview. Secondary analysis of data was conducted on a prospective longitudinal study of subjects who were at least 18 years of age and were recipients of liver and/or kidney transplant at Mayo Clinic Scottsdale, Arizona. Methods. DNA extractions of blinded urine and blood specimens of transplant patients collected at Mayo Clinic during routine transplant patient visits were performed at Baylor College of Medicine using Qiagen kits. Virologic assays included testing DNA samples for specific polyomavirus sequences using QPCR technology. De-identified demographic and clinical patient data were merged with laboratory data and statistical analysis was performed using Stata10. Results. 76 patients enrolled in the study were followed for 3.9 years post transplantation. The prevalence of BK virus and JC virus urinary excretion was 30% and 28%. Significant association was observed between JC virus excretion and kidney as the transplanted organ (P = 0.039, Pearson Chi-square test). The median urinary JCV viral loads were two logs higher than those of BKV. Patients that excreted both BKV and JCV appeared to have the worst renal function with a mean creatinine clearance value of 71.6 millimeters per minute. A survival disadvantage was observed for dual shedders of BKV and JCV, log-rank statistics, p = 0.09; 2/5 dual-shedders expired during the study period. Liver transplant and male sex were determined to be potential risk factors for JC virus activation in renal and liver transplant recipients. All patients tested negative for SV40 and no association was observed between polyomavirus excretion and type of immunosuppressive medication (tacrolimus, mycophenolate mofetil, cyclosporine and sirolimus). Conclusions. Polyomavirus reactivation was common after solid-organ transplantation and may be associated with impaired renal function. Male sex and JCV infection may be potential risk factors for viral reactivation; findings should be confirmed in larger studies.^

Decomposition of $\rm R\times C$ contingency table chi -square: Application to binned DNA fragment size data and population structure analysis

The purpose of this research is to develop a new statistical method to determine the minimum set of rows (R) in a R x C contingency table of discrete data that explains the dependence of observations. The statistical power of the method will be empirically determined by computer simulation to judge its efficiency over the presently existing methods. The method will be applied to data on DNA fragment length variation at six VNTR loci in over 72 populations from five major racial groups of human (total sample size is over 15,000 individuals; each sample having at least 50 individuals). DNA fragment lengths grouped in bins will form the basis of studying inter-population DNA variation within the racial groups are significant, will provide a rigorous re-binning procedure for forensic computation of DNA profile frequencies that takes into account intra-racial DNA variation among populations. ^