Search

Given the growing number of prediction algorithms developed to predict COVID-19 mortality, we evaluated the transportability of a mortality prediction algorithm using a multi-national network of healthcare systems. We predicted COVID-19 mortality using baseline commonly measured laboratory values and standard demographic and clinical covariates across healthcare systems, countries, and continents. Specifically, we trained a Cox regression model with nine measured laboratory test values, standard demographics at admission, and comorbidity burden pre-admission. These models were compared at site, country, and continent level. Of the 39,969 hospitalized patients with COVID-19 (68.6% male), 5717 (14.3%) died. In the Cox model, age, albumin, AST, creatine, CRP, and white blood cell count are most predictive of mortality. The baseline covariates are more predictive of mortality during the early days of COVID-19 hospitalization. Models trained at healthcare systems with larger cohort size largely retain good transportability performance when porting to different sites. The combination of routine laboratory test values at admission along with basic demographic features can predict mortality in patients hospitalized with COVID-19. Importantly, this potentially deployable model differs from prior work by demonstrating not only consistent performance but also reliable transportability across healthcare systems in the US and Europe, highlighting the generalizability of this model and the overall approach.

Modeling disease transmission is important yet challenging during a pandemic. It is not only critical to study the generative nature of a virus and understand the dynamics of transmission that vary by time and location but also crucial to provide accurate estimates that are robust to the data errors and model assumptions due to limited understanding of the new virus. We bridged the epidemiology of pathogens and statistical techniques to propose a hybrid statistical-epidemiological model so that the model is generative in nature yet utilizes statistics to model the effect of covariates (e.g., time and interventions), stochastic dependence, and uncertainty of data. Our model considers disease case counts as random variables and imposes moment assumptions on the number of secondary cases of existing infected individuals at the time since infection. Under the quasi-score framework, we built an "observation-driven model" to model the serial dependence and estimate the effect of covariates, with the capability of accounting for errors in the data. We proposed an online estimator and an effective iterative algorithm to estimate the instantaneous reproduction number and covariate effects, which provide a close monitoring and dynamic forecasting of disease transmission under different conditions to support policymaking at the community level.

We develop a Bayesian graphical modeling framework for functional data for correlated multivariate random variables observed over a continuous domain. Our method leads to graphical Markov models for functional data which allows the graphs to vary over the functional domain. The model involves estimation of graphical models that evolve functionally in a nonparametric fashion while accounting for within-functional correlations and borrowing strength across functional positions so contiguous locations are encouraged but not forced to have similar graph structure and edge strength. We utilize a strategy that combines nonparametric basis function modeling with modified Bayesian graphical regularization techniques, which induces a new class of hypoexponential normal scale mixture distributions that not only leads to adaptively shrunken estimators of the conditional cross-covariance but also facilitates a thorough theoretical investigation of the shrinkage properties. Our approach scales up to large functional datasets collected on a fine grid. We show through simulations and real data analysis that the Bayesian functional graphical model can efficiently reconstruct the functionally-evolving graphical models by accounting for within-function correlations.

Patient-derived xenografts (PDX) are tumor-in-mouse models for cancer. PDX collections, such as the NCI PDXNet, are powerful resources for preclinical therapeutic testing. However, variations in experimental and analysis procedures have limited interpretability. To determine the robustness of PDX studies, the PDXNet tested temozolomide drug response for three prevalidated PDX models (sensitive, resistant, and intermediate) across four blinded PDX Development and Trial Centers using independently selected standard operating procedures. Each PDTC was able to correctly identify the sensitive, resistant, and intermediate models, and statistical evaluations were concordant across all groups. We also developed and benchmarked optimized PDX informatics pipelines, and these yielded robust assessments across xenograft biological replicates. These studies show that PDX drug responses and sequence results are reproducible across diverse experimental protocols. In addition, we share the range of experimental procedures that maintained robustness, as well as standardized cloud-based workflows for PDX exome-sequencing and RNA-sequencing analyses and for evaluating growth. SIGNIFICANCE: The PDXNet Consortium shows that PDX drug responses and sequencing results are reproducible across diverse experimental protocols, establishing the potential for multisite preclinical studies to translate into clinical trials.

Functional quantile regression (FQR) is a useful alternative to mean regression for functional data as it provides a comprehensive understanding of how scalar predictors influence the conditional distribution of functional responses. In this article, we study the FQR model for densely sampled, high-dimensional functional data without relying on parametric or independent assumptions on the residual process, with the focus on statistical inference and scalable implementation. This is achieved by a simple but powerful distributed strategy, in which we first perform separate quantile regression to compute $M$-estimators at each sampling location, and then carry out estimation and inference for the entire coefficient functions by properly exploiting the uncertainty quantification and dependence structure of $M$-estimators. We derive a uniform Bahadur representation and a strong Gaussian approximation result for the $M$-estimators on the discrete sampling grid, serving as the basis for inference. An interpolation-based estimator with minimax optimality is proposed, and large sample properties for point and simultaneous interval estimators are established. The obtained minimax optimal rate under the FQR model shows an interesting phase transition phenomenon that has been previously observed in functional mean regression. The proposed methods are illustrated via simulations and an application to a mass spectrometry proteomics dataset.

In this paper, we develop a quantile functional regression modeling framework that models the distribution of a set of common repeated observations from a subject through the quantile function, which is regressed on a set of covariates to determine how these factors affect various aspects of the underlying subject-specific distribution. To account for smoothness in the quantile functions, we introduce custom basis functions we call \textit{quantlets} that are sparse, regularized, near-lossless, and empirically defined, adapting to the features of a given data set and containing a Gaussian subspace so {non-Gaussianness} can be assessed. While these quantlets could be used within various functional regression frameworks, we build a Bayesian framework that uses nonlinear shrinkage of quantlet coefficients to regularize the functional regression coefficients and allows fully Bayesian inferences after fitting a Markov chain Monte Carlo. Specifically, we apply global tests to assess which covariates have any effect on the distribution at all, followed by local tests to identify at which specific quantiles the differences lie while adjusting for multiple testing, and to assess whether the covariate affects certain major aspects of the distribution, including location, scale, skewness, Gaussianness, or tails. If the difference lies in these commonly-used summaries, our approach can still detect them, but our systematic modeling strategy can also detect effects on other aspects of the distribution that might be missed if one restricted attention to pre-chosen summaries. We demonstrate the benefit of the basis space modeling through simulation studies, and illustrate the method using a biomedical imaging data set in which we relate the distribution of pixel intensities from a tumor image to various demographic, clinical, and genetic characteristics., Comment: 41 pages, 8 figures