Variable selection methods have been extensively developed for and applied to cancer genomics data to identify important omics features associated with complex disease traits, including cancer outcomes. However, the reliability and reproducibility of the findings are in question if valid inferential procedures are not available to quantify the uncertainty of the findings. In this article, we provide a gentle but systematic review of high-dimensional frequentist and Bayesian inferential tools under sparse models which can yield uncertainty quantification measures, including confidence (or Bayesian credible) intervals, p values and false discovery rates (FDR). Connections in high-dimensional inferences between the two realms have been fully exploited under the “unpenalized loss function + penalty term” formulation for regularization methods and the “likelihood function × shrinkage prior” framework for regularized Bayesian analysis. In particular, we advocate for robust Bayesian variable selection in cancer genomics studies due to its ability to accommodate disease heterogeneity in the form of heavy-tailed errors and structured sparsity while providing valid statistical inference. The numerical results show that robust Bayesian analysis incorporating exact sparsity has yielded not only superior estimation and identification results but also valid Bayesian credible intervals under nominal coverage probabilities compared with alternative methods, especially in the presence of heavy-tailed model errors and outliers.

Gut microbiota have been implicated in severe malaria in murine models, but their contribution to the pathogenesis of severe malaria in children is unknown. Here we show through analysis of gut bacteria in stool samples from two separate African studies enrolling children with severe malaria, and children from local communities, that children with severe malaria have gut bacteria dysbiosis. Among children with severe malaria, there is increased abundance of Enterobacteriaceae that associates with multiple clinical complications of severe malaria. Moreover, increased abundance of Escherichia coli was a predictor of post-discharge mortality. Metagenome analysis identify elevated metabolic pathways and genes supporting the utilization of host-derived molecules in children with severe malaria that have the potential to promote the survival and growth of Enterobacteriaceae. Treatments that target Enterobacteriaceae may have the potential to reduce post-discharge mortality in children with severe malaria.

We introduce interep, an R package for interaction analysis of repeated measurement data with high-dimensional main and interaction effects. In G × E interaction studies, the forms of environmental factors play a critical role in determining how structured sparsity should be imposed in the high-dimensional scenario to identify important effects. Zhou et al. (2019) (PMID: 31816972) proposed a longitudinal penalization method to select main and interaction effects corresponding to the individual and group structure, respectively, which requires a mixture of individual and group level penalties. The R package interep implements generalized estimating equation (GEE)-based penalization methods with this sparsity assumption. Moreover, alternative methods have also been implemented in the package. These alternative methods merely select effects on an individual level and ignore the group-level interaction structure. In this software article, we first introduce the statistical methodology corresponding to the penalized GEE methods implemented in the package. Next, we present the usage of the core and supporting functions, which is followed by a simulation example with R codes and annotations. The R package interep is available at The Comprehensive R Archive Network (CRAN).

Background: Molecular-based approaches to understanding concussion pathophysiology provide complex biological information that can advance concussion research and identify potential diagnostic and/or prognostic biomarkers of injury. Objective: The aim of this study was to identify gene expression changes in peripheral blood that are initiated following concussion and are relevant to concussion response and recovery. Methods: We analyzed whole blood transcriptomes in a large cohort of concussed and control collegiate athletes who were participating in the multicenter prospective cohort Concussion Assessment, Research, and Education (CARE) Consortium study. Blood samples were collected from collegiate athletes at preseason (baseline), within 6 h of concussion injury, and at four additional prescribed time points spanning 24 h to 6 months post-injury. RNA sequencing was performed on samples from 230 concussed, 130 contact control, and 102 non-contact control athletes. Differential gene expression and deconvolution analysis were performed at each time point relative to baseline. Results: Cytokine and immune response signaling pathways were activated immediately after concussion, but at later time points these pathways appeared to be suppressed relative to the contact control group. We also found that the proportion of neutrophils increased and natural killer cells decreased in the blood following concussion. Conclusions: Transcriptome signatures in the blood reflect the known pathophysiology of concussion and may be useful for defining the immediate biological response and the time course for recovery. In addition, the identified immune response pathways and changes in immune cell type proportions following a concussion may inform future treatment strategies.

Paediatric acute kidney injury (AKI) is a complication of severe malaria, but its long-term outcomes remain poorly defined in low-income and middle-income countries. We aimed to evaluate the long-term association between paediatric AKI and kidney outcomes and mortality following severe malaria. We pooled data from two prospective cohorts of Ugandan children (ie, those aged 6 months-12 years) admitted to hospital with severe malaria between 2008 and 2017. Children with stored admission blood samples available for creatinine measurement were included in the analysis. Surviving participants were enrolled in a follow-up study conducted from 2020 to 2023, when kidney function and survival were assessed, using Cox regression with age as the timescale to model mortality risk. Logistic regression was used to estimate the odds of chronic kidney disease (CKD), defined as two or more consecutive low estimated glomerular filtration rates 90 days or more apart, and long-term major adverse kidney events, defined as CKD or death. Adjusted analyses included the following enrolment characteristics: age, sex, height-for-age Z score, study site, study cohort, severe malaria group (ie, cerebral malaria, severe malarial anaemia, respiratory distress, complicated seizures, or prostration), and HIV status. At enrolment, the median age was 2·5 years (IQR 1·8-3·5), 622 (57·8%) of 1077 were male and 455 (42·2%) were female, with Kidney Disease: Improving Global Outcomes-defined AKI occurring in 431 (40·0%) children. Over a median of 6·5 years (3·7-8·8), 147 (13·6%) of children died, with nearly half of deaths (71 of 147) occurring after hospital discharge. Adjusted odds of long-term major adverse kidney events were higher among children with AKI (adjusted odds ratio [aOR] 3·14, 95% CI 2·23-4·43), driven by a higher mortality risk (adjusted hazard ratio [aHR] 3·36, 95% CI 2·22-5·10) that remained elevated beyond 2 years after the acute episode (aHR 4·53, 1·80-11·37). AKI survivors had higher odds of chronic kidney disease over follow-up (odds ratio 1·77, 1·07-2·93), although not significant after adjustment (aOR 1·47, 95% CI 0·84-2·56; p=0·18). In this paediatric population, AKI was associated with excess long-term mortality, suggesting AKI is a sentinel event that marks sustained vulnerability to death and highlights limitations of acute-care models in malaria-endemic settings. The US National Institute of Neurological Disorders and Stroke, the Fogarty International Center, the US National Institutes of Allergy and Infectious Diseases, and a Ralph W and Grace M Showalter Young Investigator Award.

Sport-related head exposure characteristics (i.e., concussion history, sport contact classification, and total years of sport participation) are suggested to be associated with short-term and long-term health outcomes in competitive athletes. However, their relationship with intermediate health outcomes (specifically within 5 years of collegiate sport retirement) remains unknown. Therefore, we examined associations between sport-related head exposure characteristics and physical, mental, cognitive, and behavioral health measures among former collegiate athletes. Former collegiate athletes who completed a baseline evaluation between 2018 and 2021 and were evaluated within 5 years of collegiate graduation were included. Primary predictors included lifetime concussion history (0 [referent], 1-2, 3+), sport contact classification (unexposed [ref.], low, high), and total years of sport participation (continuous). Outcome measures collected included the Alcohol Use Disorders Identification Test, Brief Symptom Inventory-18 (BSI-18), Neuro-Quality of Life cognitive domain (Neuro-QoL), Patient Health Questionnaire (PHQ-9), Pittsburgh Sleep Quality Index (PSQI), Sport Concussion Assessment Tool (SCAT) symptom inventory, 12-Item Short Form Survey (SF-12), and Satisfaction with Life Scale. Twelve multivariable linear regression models with bootstrapping were constructed, with adjusted unstandardized beta coefficients (B_Adj) and standard errors (SEs) estimated. Overall, 3,910 former collegiate student-athletes were examined; 48.6% were female, 73.0% attended a Division I school, 36.2% self-reported ≥1 lifetime diagnosed concussion, 38.1% participated in a high-exposure sport, and the median years since college graduation to assessment was 5.0. Independently, scores of BSI anxiety (B_Adj = 1.24, SE = 0.36), BSI depression (B_Adj = 0.88, SE = 0.37), global severity index [GSI] (B_Adj = 2.66, SE = 0.91), PHQ-9 (B_Adj = 1.34, SE = 0.40), PSQI (B_Adj = 0.69, SE = 0.29), SCAT total symptom severity (B_Adj = 5.14, SE = 1.31), and SF-12 mental composite score (B_Adj = -3.39, SE = 0.83) were worse in athletes with 3+ concussions compared with athletes without a history of concussion. Athletes with 1-2 concussions vs no concussion history similarly had worse scores independently for BSI-18 GSI (B_Adj = 0.88, SE = 0.35), anxiety (B_Adj = 0.40, SE = 0.15), Neuro-QoL (B_Adj = -0.97, SE = 0.36), PHQ-9 (B_Adj = 0.77, SE = 0.19), PSQI (B_Adj = 0.27, SE = 0.13), SCAT total symptom severity (B_Adj = 2.13, SE = 0.55), and SF-12 mental composite score (B_Adj = -1.80, SE = 0.41). Lifetime concussion history was the most common predictor of adverse self-reported health outcomes compared with other exposure characteristics, within 5 years of collegiate sport retirement. Despite the associations, most former athletes remained within normal clinical levels during this early postretirement period.