Biometrics Section Early Career Paper Award Session

Estimation of covariance or inverse covariance matrices is fundamental across numerous scientific fields. Recently, increasing attention has been directed toward incorporating covariate effects into these matrices, facilitating subject-specific estimation. Despite these advances, guaranteeing the positive definiteness of the resulting estimators remains a challenging problem. In this paper, we present a new varying-coefficient sequential regression framework that extends the modified Cholesky decomposition to model the positive definite covariance matrix as a function of subject-level covariates. To handle high-dimensional responses and covariates, we impose a joint sparsity structure that simultaneously promotes sparsity in both the covariate effects and the entries in the Cholesky factors that are modulated by these covariates. We approach parameter estimation with a blockwise coordinate descent algorithm, and investigate the ℓ₂ convergence rate of the estimated parameters. The efficacy of the proposed method is demonstrated through numerical experiments and an application to a gene co-expression network study with brain cancer patients to determine if and how gene co-expressions vary with genetic variations. 

We consider parameter estimation and inference when data feature blockwise, nonmonotone missingness. Our approach, rooted in semiparametric theory and inspired by prediction-powered inference, leverages off-the-shelf AI (predictive or generative) models to handle missing completely at random mechanisms, by finding an approximation of the optimal estimating equation through a novel and tractable Restricted Anova hierarchY (RAY) approximation. The resulting Inference for Blockwise Missingness (RAY), or IBM(RAY) estimator incorporates pre-trained AI models and carefully controls asymptotic variance by tuning model-specific hyperparameters. We then extend IBM(RAY) to a general class of estimators. We find the most efficient estimator in this class, which we call IBM(Adaptive), by solving a constrained quadratic programming problem. All IBM estimators are unbiased, and, crucially, asymptotically achieving guaranteed efficiency gains over a naive complete-case estimator, regardless of the predictive accuracy of the AI models used. We demonstrate the finite-sample performance and numerical stability of our method through simulation studies and an application to surface protein abundance estimation. 

Understanding how vaccine effectiveness (VE) changes over time can provide evidence-based guidance
for public health decision making. While commonly reported by practitioners, time-varying VE estimates obtained
using Cox regression are vulnerable to hidden biases. To address these limitations, we describe how to leverage
vaccine-irrelevant infections to identify hazard-based, time-varying VE in the presence of unmeasured confounding
and selection bias. We articulate assumptions under which our approach identifies a causal effect of an intervention
deferring vaccination and interaction with the community in which infections circulate. We develop sieve and efficient influence curve-based estimators and discuss imposing monotone shape constraints and estimating VE against multiple variants. As a case study, we examine the observational booster phase of the Coronavirus Vaccine Efficacy (COVE) trial of the Moderna mRNA-1273 COVID-19 vaccine which used symptom-triggered multiplex PCR testing to identify acute respiratory illnesses (ARIs) caused by SARS-CoV-2 and 20 off-target pathogens previously identified as compelling negative controls for COVID-19. Accounting for vaccine-irrelevant ARIs supported that the mRNA-1273 booster was more effective and durable against Omicron COVID-19 than suggested by Cox regression. Our work offers an approach to mitigate bias in hazard-based, time-varying treatment effects in randomized and non-randomized studies using negative controls. 

An increasing number of microbiome studies are simultaneously collecting host omics profiles to better understand microbe-host interplay. Although, there remain analytic challenges due to the dimensionality, heterogeneity, and sparsity of the data, as well as the complexity of interactions within and between modalities. We focus on measuring global multivariate associations between the microbiome and a host omics modality, as it is often the first step in multiomics analysis due to its ability to aggregate effects across all features. Additionally, in microbiome studies global analysis has biologically meaningful connections to ecological diversity. We propose a distance-based mutual information test of global association for microbiome multiomics integration. The use of mutual information is advantageous, as it is able to capture general forms of dependence. This includes linear, nonlinear, monotone, and nonmonotone associations, which most correlation metrics can fail to capture. We develop a k-nearest neighbor estimation procedure and corresponding permutation test for independence, as well as an ensemble distance metric. Simulation studies show that our distance-based mutual information test controls type I error and maintains good power under a variety of alternative dependence structures. We apply our method to a multiple sclerosis case-control study to elucidate the gut microbiome's association with stool and plasma short chain fatty acids.