Search

Motivated by image‐on‐scalar regression with data aggregated across multiple sites, we consider a setting in which multiple independent studies each collect multiple dependent vector outcomes, with potential mean model parameter homogeneity between studies and outcome vectors. To determine the validity of a joint analysis of these data sources, we must learn which of them share mean model parameters. We propose a new model fusion approach that delivers improved flexibility and statistical performance over existing methods. Our proposed approach specifies a quadratic inference function within each data source and fuses mean model parameter vectors in their entirety based on a new formulation of a pairwise fusion penalty. We establish theoretical properties of our estimator and propose an asymptotically equivalent weighted oracle meta‐estimator that is more computationally efficient. Simulations and an application to the ABIDE neuroimaging consortium highlight the flexibility of the proposed approach. An R package is provided for ease of implementation. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this study was to determine the association between prenatal and early life exposure to lead and the presence of molar hypomineralization (MH) in a group of Mexican children. Methods: A subset of participants of the Early Life Exposure in Mexico to Environmental Toxicants (ELEMENTS) cohort study was examined for the presence of molar hypomineralization using European Academy of Pediatric Dentistry (EAPD) criteria. Prenatal lead exposure was assessed by K-ray fluorescence measurements of patella and tibia lead and by maternal blood lead levels by trimester and averaged over trimesters. Postnatal exposure was assessed by levels of maternal blood lead at delivery and child blood lead at 12 and 24 months. Results: A subset of 506 subjects from the ELEMENT cohorts (nine to 18 years old) were examined for MH; 87 subjects (17.2 percent) had MH. Maternal blood lead levels in the third trimester (odds ratio [OR] equals 1.08; 95 percent confidence interval [95% CI] equals 1.02 to 1.15) and averaged over three trimesters (OR equals 1.10; 95% CI equals 1.02 to 1.19) were significantly associated with MH status. None of the maternal bone lead or the child’s blood lead parameters was significantly associated with the presence of MH (P>0.05). Conclusions: This study documents a significant association between prenatal lead exposure especially in late pregnancy and the odds of molar hypomineralization. [ABSTRACT FROM AUTHOR]

Abstract Extreme environmental events frequently exhibit spatial and temporal dependence. These data are often modeled using Max Stable Processes (MSPs) that are computationally prohibitive to fit for as few as a dozen observations. Supposed computationally-efficient approaches like the composite likelihood remain computationally burdensome with a few hundred observations. In this article, we propose a spatial partitioning approach based on local modeling of subsets of the spatial domain that delivers computationally and statistically efficient inference. Marginal and dependence parameters of the MSP are estimated locally on subsets of observations using censored pairwise composite likelihood, and combined using a modified generalized method of moments procedure. The proposed distributed approach is extended to estimate inverted MSP models, and to estimate spatially varying coefficient models to deliver computationally efficient modeling of spatial variation in marginal parameters. We demonstrate consistency and asymptotic normality of estimators, and show empirically that our approach leads to statistically efficient estimation of model parameters. We illustrate the flexibility and practicability of our approach through simulations and the analysis of streamflow data from the U.S. Geological Survey. Supplementary materials for this article are available online. [ABSTRACT FROM AUTHOR]

This article is motivated by a regression analysis of electroencephalography (EEG) neuroimaging data with high-dimensional correlated responses with multilevel nested correlations. We develop a divide-and-conquer procedure implemented in a fully distributed and parallelized computational scheme for statistical estimation and inference of regression parameters. Despite significant efforts in the literature, the computational bottleneck associated with high-dimensional likelihoods prevents the scalability of existing methods. The proposed method addresses this challenge by dividing responses into subvectors to be analyzed separately and in parallel on a distributed platform using pairwise composite likelihood. Theoretical challenges related to combining results from dependent data are overcome in a statistically efficient way using a meta-estimator derived from Hansen's generalized method of moments. We provide a rigorous theoretical framework for efficient estimation, inference, and goodness-of-fit tests. We develop an R package for ease of implementation. We illustrate our method's performance with simulations and the analysis of the EEG data, and find that iron deficiency is significantly associated with two auditory recognition memory related potentials in the left parietal-occipital region of the brain. for this article are available online. [ABSTRACT FROM AUTHOR]

Epigenetic modifications, such as DNA methylation, influence gene expression and cardiometabolic phenotypes that are manifest in developmental periods in later life, including adolescence. Untargeted metabolomics analysis provide a comprehensive snapshot of physiological processes and metabolism and have been related to DNA methylation in adults, offering insights into the regulatory networks that influence cellular processes. We analyzed the cross-sectional correlation of blood leukocyte DNA methylation with 3758 serum metabolite features (574 of which are identifiable) in 238 children (ages 8-14 years) from the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) study. Associations between these features and percent DNA methylation in adolescent blood leukocytes at LINE-1 repetitive elements and genes that regulate early life growth (IGF2, H19, HSD11B2) were assessed by mixed effects models, adjusting for sex, age, and puberty status. After false discovery rate correction (FDR q < 0.05), 76 metabolites were significantly associated with LINE-1 DNA methylation, 27 with HSD11B2, 103 with H19, and 4 with IGF2. The ten identifiable metabolites included dicarboxylic fatty acids (five associated with LINE-1 or H19 methylation at q < 0.05) and 1-octadecanoyl-rac-glycerol (q < 0.0001 for association with H19 and q = 0.04 for association with LINE-1). We then assessed the association between these ten known metabolites and adiposity 3 years later. Two metabolites, dicarboxylic fatty acid 17:3 and 5-oxo-7-octenoic acid, were inversely associated with measures of adiposity (P <.05) assessed approximately 3 years later in adolescence. In stratified analyses, sex-specific and puberty-stage specific (Tanner stage = 2 to 5 vs Tanner stage = 1) associations were observed. Most notably, hundreds of statistically significant associations were observed between H19 and LINE-1 DNA methylation and metabolites among children who had initiated puberty. Understanding relationships between subclinical molecular biomarkers (DNA methylation and metabolites) may increase our understanding of genes and biological pathways contributing to metabolic changes that underlie the development of adiposity during adolescence. [ABSTRACT FROM AUTHOR]

This paper presents a unified framework for supervised learning and inference procedures using the divide-and-conquer approach for high-dimensional correlated outcomes. We propose a general class of estimators that can be implemented in a fully distributed and parallelized computational scheme. Modeling, computational and theoretical challenges related to high-dimensional correlated outcomes are overcome by dividing data at both outcome and subject levels, estimating the parameter of interest from blocks of data using a broad class of supervised learning procedures, and combining block estimators in a closed-form meta-estimator asymptotically equivalent to estimates obtained by Hansen (1982)'s generalized method of moments (GMM) that does not require the entire data to be reloaded on a common server. We provide rigorous theoretical justifications for the use of distributed estimators with correlated outcomes by studying the asymptotic behaviour of the combined estimator with fixed and diverging number of data divisions. Simulations illustrate the finite sample performance of the proposed method, and we provide an R package for ease of implementation. [ABSTRACT FROM AUTHOR]

Epigenetic modifications, such as DNA methylation, influence gene expression and cardiometabolic phenotypes that are manifest in developmental periods in later life, including adolescence. Untargeted metabolomics analysis provide a comprehensive snapshot of physiological processes and metabolism and have been related to DNA methylation in adults, offering insights into the regulatory networks that influence cellular processes. We analyzed the cross-sectional correlation of blood leukocyte DNA methylation with 3758 serum metabolite features (574 of which are identifiable) in 238 children (ages 8-14 years) from the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) study. Associations between these features and percent DNA methylation in adolescent blood leukocytes at LINE-1 repetitive elements and genes that regulate early life growth (IGF2, H19, HSD11B2) were assessed by mixed effects models, adjusting for sex, age, and puberty status. After false discovery rate correction (FDR q < 0.05), 76 metabolites were significantly associated with LINE-1 DNA methylation, 27 with HSD11B2, 103 with H19, and 4 with IGF2. The ten identifiable metabolites included dicarboxylic fatty acids (five associated with LINE-1 or H19 methylation at q < 0.05) and 1-octadecanoyl-rac-glycerol (q < 0.0001 for association with H19 and q = 0.04 for association with LINE-1). We then assessed the association between these ten known metabolites and adiposity 3 years later. Two metabolites, dicarboxylic fatty acid 17:3 and 5-oxo-7-octenoic acid, were inversely associated with measures of adiposity (P <.05) assessed approximately 3 years later in adolescence. In stratified analyses, sex-specific and puberty-stage specific (Tanner stage = 2 to 5 vs Tanner stage = 1) associations were observed. Most notably, hundreds of statistically significant associations were observed between H19 and LINE-1 DNA methylation and metabolites among children who had initiated puberty. Understanding relationships between subclinical molecular biomarkers (DNA methylation and metabolites) may increase our understanding of genes and biological pathways contributing to metabolic changes that underlie the development of adiposity during adolescence. [ABSTRACT FROM AUTHOR]

This document is a response to a discussion on statistical inference for streamed longitudinal data. The authors address three important points raised by the discussants: the role of the weighting parameter in balancing bias and variance, the use of subsampling for computational efficiency, and the extension to high-dimensional settings. They provide insights and evidence on the robustness of their estimator to low data contamination and the impact of the weighting parameter on bias and variance trade-off. The authors also discuss the potential benefits of subsampling in reducing correlation among selected subsamples. They conducted numerical experiments comparing the performance of ordinary least-squares and weighted least-squares estimators, finding that the ability to ignore correlation depends on the true correlation coefficient and sampling frequency. The authors caution that the choice of sampling frequency should be made carefully, as it depends on the underlying strength of correlation. They also propose using the stochastic gradient descent algorithm for more efficient estimation and inference in high-dimensional settings. [Extracted from the article]

Modern longitudinal data, for example from wearable devices, may consist of measurements of biological signals on a fixed set of participants at a diverging number of time-points. Traditional statistical methods are not equipped to handle the computational burden of repeatedly analysing the cumulatively growing dataset each time new data are collected. We propose a new estimation and inference framework for dynamic updating of point estimates and their standard errors along sequentially collected datasets with dependence, both within and between the datasets. The key technique is a decomposition of the extended inference function vector of the quadratic inference function constructed over the cumulative longitudinal data into a sum of summary statistics over data batches. We show how this sum can be recursively updated without the need to access the whole dataset, resulting in a computationally efficient streaming procedure with minimal loss of statistical efficiency. We prove consistency and asymptotic normality of our streaming estimator as the number of data batches diverges, even as the number of independent participants remains fixed. Simulations demonstrate the advantages of our approach over traditional statistical methods that assume independence between data batches. Finally, we investigate the relationship between physical activity and several diseases through analysis of accelerometry data from the National Health and Nutrition Examination Survey. [ABSTRACT FROM AUTHOR]

Rationale: Mass spectrometry imaging (MSI) elevates the power of conventional mass spectrometry (MS) to multidimensional space, elucidating both chemical composition and localization. However, the field lacks any robust quality control (QC) and/or system suitability testing (SST) protocols to monitor inconsistencies during data acquisition, both of which are integral to ensure the validity of experimental results. To satisfy this demand in the community, we propose an adaptable QC/SST approach with five analyte options amendable to various ionization MSI platforms (e.g., desorption electrospray ionization, matrix‐assisted laser desorption/ionization [MALDI], MALDI‐2, and infrared matrix‐assisted laser desorption electrospray ionization [IR‐MALDESI]). Methods: A novel QC mix was sprayed across glass slides to collect QC/SST regions‐of‐interest (ROIs). Data were collected under optimal conditions and on a compromised instrument to construct and refine the principal component analysis (PCA) model in R. Metrics, including mass measurement accuracy and spectral accuracy, were evaluated, yielding an individual suitability score for each compound. The average of these scores is utilized to inform if troubleshooting is necessary. Results: The PCA‐based SST model was applied to data collected when the instrument was compromised. The resultant SST scores were used to determine a statistically significant threshold, which was defined as 0.93 for IR‐MALDESI‐MSI analyses. This minimizes the type‐I error rate, where the QC/SST would report the platform to be in working condition when cleaning is actually necessary. Further, data scored after a partial cleaning demonstrate the importance of QC and frequent full instrument cleaning. Conclusions: This study is the starting point for addressing an important issue and will undergo future development to improve the efficiency of the protocol. Ultimately, this work is the first of its kind and proposes this approach as a proof of concept to develop and implement universal QC/SST protocols for a variety of MSI platforms. [ABSTRACT FROM AUTHOR]

Mercury intoxication is known to be associated with adverse symptoms of fatigue and sleep disturbances, but whether low-level mercury exposure could affect sleep remains unclear. In particular, children may be especially vulnerable to both mercury exposures and to poor sleep. We sought to examine associations between mercury levels and sleep disturbances in Mexican youth. The study sample comprised 372 youth from the Early Life Exposures to Environmental Toxicants (ELEMENT) cohort, a birth cohort from Mexico City. Sleep (via 7-day actigraphy) and concurrent urine mercury were assessed during a 2015 follow-up visit. Mercury was also assessed in mid-childhood hair, blood, and urine during an earlier study visit, and was considered a secondary analysis. We used linear regression and varying coefficient models to examine non-linear associations between Hg exposure biomarkers and sleep duration, timing, and fragmentation. Unstratified and sex-stratified analyses were adjusted for age and maternal education. During the 2015 visit, participants were 13.3 ± 1.9 years, and 48% were male. There was not a cross-sectional association between urine Hg and sleep characteristics. In secondary analysis using earlier biomarkers of Hg, lower and higher blood Hg exposure was associated with longer sleep duration among girls only. In both boys and girls, Hg biomarker levels in 2008 were associated with later adolescent sleep midpoint (for Hg urine in girls, and for blood Hg in boys). For girls, each unit log Hg was associated with 0.2 h later midpoint (95% CI 0 to 0.4), and for boys each unit log Hg was associated with a 0.4 h later sleep midpoint (95% CI 0.1 to 0.8). There were mostly null associations between Hg exposure and sleep characteristics among Mexican children. Yet, in both boys and girls, higher Hg exposure in mid-childhood (measured in urine and blood, respectively) was related to later sleep timing in adolescence. • Whether low-level mercury exposure is related to poor sleep is an underexplored research area. • In cross-sectional analyses, urine Hg was not related to sleep in adolescents. • Yet, prospective analysis provided some evidence that Hg may be related to sleep. • Prospective positive associations between Hg and sleep duration among girls were found. • Prospective associations between Hg and later sleep timing for both boys and girls were noted. [ABSTRACT FROM AUTHOR]