Substantially motivated by analytic needs from a NIH/EPA co-funded P01 project, this project develops a set of novel statistical models and efficient algorithms to evaluate, interpret, and predict impacts of early life exposures to environmental risk factors on children health and developmental outcomes. The P01 project has collected massive complex data from multiple longitudinal birth cohorts for a systematic study of the hypothesis that environmental exposures (e.g. BPA, phthalates and heavy metals) and during pre- and/or post-natal life influence developmental plasticity, thereby altering susceptibility to adult chronic diseases. This project will be accomplished through the following specific aims:
Aim 1 : Develop statistical models and algorithms to evaluate delayed somatic growth and sexual maturation during the adolescent period driven by prenatal and/or postnatal exposures. These methods will help us assess if, and which toxicant sets, and in which form, modify children growth patterns and tempo of sexual maturation.
Aim 2 : Develop semi-parametric stochastic models to evaluate the functional rate of growth changes and its potential exposure-driven alterations during the child period of 0-5 years old. This new class of stochastic differential equations allows analyze both growth velocity and acceleration as functions of subject's characteristics and environmental exposures to better understand, evaluate and predict child growth patterns.
Aim 3 : Develop statistical methods to assess the validity of merging longitudinal cohort data and to perform joint analysis of merged longitudinal data. Merging longitudinal data sets from multiple cohorts is complicated by the underlying heterogeneity across cohorts, and proper methods are needed to validate the data merging as well as to perform a valid joint analysis of merged data.
Aim 4 : Develop, test, distribute, and support freely available implementations of the proposed methods in this application. The developed statistical tools can be applied to analyze data form longitudinal cohort studies of similar scale and complexity. The advanced and expanded statistical models and methods can provide adequate answers to the most benefit of disease prevention and health practice as well the best interest of well beings.

Public Health Relevance

Emerging areas of research in environmental health sciences (EHS) have led to the development of new hypotheses and technologies to collect massive complex data in the EHS that present great challenges to the data analysis. The existing statistical methodologies are not inadequate to address several major analytic challenges that complicate the data analysis in many EHS studies. Substantially motivated by analytic needs from the ELEMENT projects, the primary goal of this project is to develop a set of novel statistical models and efficient algorithms to evaluate, interpret, and predict impacts of early lie exposures to environmental risk factors to children health and developmental outcomes. This new toolbox will advance and expand the statistical models and methods in EHS that provide adequate answers to the most benefit of disease prevention and health practice as well the best interest of well beings.

Agency
National Institute of Health (NIH)
Institute
National Institute of Environmental Health Sciences (NIEHS)
Type
Research Project (R01)
Project #
5R01ES024732-04
Application #
9520097
Study Section
Biostatistical Methods and Research Design Study Section (BMRD)
Program Officer
Boyles, Abee
Project Start
2015-09-01
Project End
2019-06-30
Budget Start
2018-07-01
Budget End
2019-06-30
Support Year
4
Fiscal Year
2018
Total Cost
Indirect Cost
Name
University of Michigan Ann Arbor
Department
Biostatistics & Other Math Sci
Type
Schools of Public Health
DUNS #
073133571
City
Ann Arbor
State
MI
Country
United States
Zip Code
48109
Li, Yun; Wang, Sijian; Song, Peter X-K et al. (2018) Doubly regularized estimation and selection in linear mixed-effects models for high-dimensional longitudinal data. Stat Interface 11:721-737
Perng, Wei; Baek, Jonggyu; Zhou, Christina W et al. (2018) Associations of the infancy body mass index peak with anthropometry and cardiometabolic risk in Mexican adolescents. Ann Hum Biol :1-9
Jansen, E C; Zhou, L; Song, P X K et al. (2018) Prenatal lead exposure in relation to age at menarche: results from a longitudinal study in Mexico City. J Dev Orig Health Dis 9:467-472
Zhou, Ling; Tang, Lu; Song, Angela T et al. (2017) A LASSO Method to Identify Protein Signature Predicting Post-transplant Renal Graft Survival. Stat Biosci 9:431-452
Zhou, Yan; Wang, Pei; Wang, Xianlong et al. (2017) Sparse multivariate factor analysis regression models and its applications to integrative genomics analysis. Genet Epidemiol 41:70-80
Perng, Wei; Hector, Emily C; Song, Peter X K et al. (2017) Metabolomic Determinants of Metabolic Risk in Mexican Adolescents. Obesity (Silver Spring) 25:1594-1602
Tang, Lu; Song, Peter X K (2016) Fused Lasso Approach in Regression Coefficients Clustering - Learning Parameter Heterogeneity in Data Integration. J Mach Learn Res 17:
Wang, Fei; Wang, Lu; Song, Peter X-K (2016) Fused lasso with the adaptation of parameter ordering in combining multiple studies with repeated measurements. Biometrics 72:1184-1193