一个带白噪声的可变时滞Lotka-Volterra人口模型(英文)

因为人口模型经常遭遇环境噪音的影响,本文将如下Lotka-Volterra模型(t)=diag(x(t))[b+Ax(t)+Bx(t-δ(t))]随机扰动为It型随机微分方程dx(t)=diag(x(t))[(b+Ax(t)+Bx(t-δ(t)))dt+(Qx(t)+Rx(t-δ(t))dw(t)].在这个随机模型中对系数b,A,B不需任何限制,我们证明了环境噪音不仅会压制人口的爆炸还会使得方程的解随机一致有界.     

变长度Motif识别中的Gibbs抽样算法(英文)

Motif识别是计算生物学中的重要问题.处理缺失数据的方法被大家广泛应用于生物序列中的Motif识别,例如EM算法,Gibbs抽样等等.现在识别Motif的方法都是首先假定Motif的长度是给的,但是,事实上Motif的长度是未知的,在这篇文章中,我们用Gibbs抽样算法在寻找Motif的位置的同时确定Motif的长度.     

Generalized additive modeling with implicit variable selection by likelihood-based boosting

The use of generalized additive models in statistical data analysis suffers from the restriction to few explanatory variables and the problems of selection of smoothing parameters. Generalized additive model boosting circumvents these problems by means of stagewise fitting of weak learners. A fitting procedure is derived which works for all simple exponential family distributions, including binomial, Poisson, and normal response variables. The procedure combines the selection of variables and the determination of the appropriate amount of smoothing. Penalized regression splines and the newly introduced penalized stumps are considered as weak learners. Estimates of standard deviations and stopping criteria, which are notorious problems in iterative procedures, are based on an approximate hat matrix. The method is shown to be a strong competitor to common procedures for the fitting of generalized additive models. In particular, in high-dimensional settings with many nuisance predictor variables it performs very well.     

Regularized estimation in the accelerated failure time model with high-dimensional covariates

We consider two regularization approaches, the LASSO and the threshold-gradient-directed regularization, for estimation and variable selection in the accelerated failure time model with multiple covariates based on Stute's weighted least squares method. The Stute estimator uses Kaplan-Meier weights to account for censoring in the least squares criterion. The weighted least squares objective function makes the adaptation of this approach to multiple covariate settings computationally feasible. We use V-fold cross-validation and a modified Akaike's Information Criterion for tuning parameter selection, and a bootstrap approach for variance estimation. The proposed method is evaluated using simulations and demonstrated on a real data example.     

Identity tests: determination of cell line cross-contamination

Cell line cross-contamination is a phenomenon that arises as a result of the continuous cell line culture. It has been estimated that around 20% of the cell lines are misidentified, therefore it is necessary to carry out quality control tests for the detection of this issue. Since cell line cross-contamination discovery, different methods have been applied, such as isoenzyme analysis for inter-species cross-contamination; HLA typing, and DNA fingerprinting using short tandem repeat and a variable number of tandem repeat for intra-species cross-contamination. The cell banks in this sense represent the organizations responsible for guaranteeing the authenticity of cell lines for future research and clinical uses.     

Reprint of: Neuroimaging in acute liver failure

Acute liver failure (ALF) is frequently complicated by the development of brain edema that can lead to intracranial hypertension and severe brain injury. Neuroimaging techniques allow a none-invasive assessment of brain tissue and cerebral hemodynamics by means of transcranial Doppler ultrasonography, magnetic resonance and nuclear imaging with radioligands. These methods have been very helpful to unravel the pathogenesis of this process and have been applied to patients and experimental models. They allow monitoring the outcome of patients with ALF and neurological manifestations. The increase in brain water can be detected by observing changes in brain volume and disturbances in diffusion weighted imaging. Neurometabolic changes are detected by magnetic resonance spectroscopy, which provides a pattern of abnormalities characterized by an increase in glutamine and a decrease in myo-inositol. Disturbances in cerebral blood flow are depicted by SPECT or PET and can be monitored and the bedside by assessing the characteristics of the waveform provided by transcranial Doppler ultrasonography. Neuroimaging methods, which are rapidly evolving, will undoubtedly lead to future diagnostic and therapeutic progress that could be very helpful for patients with ALF.     

LADA and T1D in Estonian population - two different genetic risk profiles

Aims/hypothesis

The aim of our study was to analyze combined impact of 17 polymorphisms at 8 gene regions previously shown to be associated with autoimmunity in diabetes. We hypothesized that the genetic predisposition is multiplicative and joint risk of different diabetic phenotypes forms by distinct combination of susceptibility loci.

Methods

An ethnically homogenous population of Estonian origin, including 65 LADA patients, 154 patients with T1D, 260 patients with T2D and 229 non-diabetic controls, was genotyped for polymorphisms/haplotypes in HLA-DQB1, insulin gene (rs689, rs3842729), PHTF1–PTPN22 region (rs2476601, rs6679677), CTLA4 region (rs231806, rs16840252, rs5742909, rs231775, rs3087243, rs2033171), ICOS region (rs10932037, rs4675379), CD25 (rs706778), CD226(rs763361), NAA25 (rs17696736).

Results

As expected, the risk of T1D was consistently attributed by HLA-DQB1 haplotypes, but also by haplotypes of INS and PHTF1–PTPN22 region, and rs17696736 in NAA25. By contrast, LADA was associated only with T1D-protective HLA haplotypes and with two more frequent haplotypes of the CTLA4. It is of interest, that seldom CT haplotype of PHTF1–PTPN22 region carried the risk for autoantibody-negative T2D. The final best-fitted model for T1D genetic risk contained six gene regions (HLA-DQB1, INS, PHTF1, CTLA4 + 49, CD226 and NAA25) and for LADA only two (HLA-DQB1 and CTLA4 + 49). The AUCs of these models are 0.869 and 0.693, respectively.

Conclusions

Classical T1D-risk haplotypes of HLA and some non-HLA loci describe quite well the genetic risk for T1D but not for LADA. The need of further studies should be stressed to discover the real risk factors for slower forms of autoimmune diabetes in adults.     

Selecting factors predictive of heterogeneity in multivariate event time data

In multivariate survival analysis, investigators are often interested in testing for heterogeneity among clusters, both overall and within specific classes. We represent different hypotheses about the heterogeneity structure using a sequence of gamma frailty models, ranging from a null model with no random effects to a full model having random effects for each class. Following a Bayesian approach, we define prior distributions for the frailty variances consisting of mixtures of point masses at zero and inverse-gamma densities. Since frailties with zero variance effectively drop out of the model, this prior allocates probability to each model in the sequence, including the overall null hypothesis of homogeneity. Using a counting process formulation, the conditional posterior distributions of the frailties and proportional hazards regression coefficients have simple forms. Posterior computation proceeds via a data augmentation Gibbs sampling algorithm, a single run of which can be used to obtain model-averaged estimates of the population parameters and posterior model probabilities for testing hypotheses about the heterogeneity structure. The methods are illustrated using data from a lung cancer trial.