Yunxuan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yunxuan Zhang has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Hongyu Zhao
Department of Biostatistics, Yale School of Public Health, New Haven, USA
3 shared publications - Emma Zang
Department of Sociology, Yale University, New Haven, USA
2 shared publications - Thomas M. Gill
Department of Internal Medicine, Yale School of Medicine , New Haven, Connecticut
2 shared publications - Yiliang Zhang
Yale University
2 shared publications - Bei Wu
NYU Shanghai, Shanghai, China
1 shared publication - Chi Zhang
Affiliation Department of Biostatistics, Yale University School of Public Health, New Haven, Connecticut, United States of America
1 shared publication
Latest Publications
- 22 September 2025
- Published by Oxford University Press (OUP) in American Journal of Epidemiology
- Vol. 195 (3) , 816-823
Obtaining population-based estimates for survey data using Bayesian hierarchical models with poststratification
For large-scale surveys such as the National Health and Aging Trends Study (NHATS), investigators may wish to combine data from two (or more) cohorts in a single analysis to obtain larger sample sizes. Unfortunately, it is not possible to combine the 2011 and 2015 NHATS cohorts while retaining the sample weights. We applied Bayesian hierarchical models with poststratification as an alternative strategy for obtaining population-based estimates from NHATS. As proof of principle, we compared prevalence estimates of frailty obtained from our Bayesian approach with those obtained from the 2011 and 2015 cohorts using the NHATS sample weights. Once validated, we applied our strategy to combine the cohorts into a single analytical dataset without overlap of participants, and generated Bayesian estimates of frailty for the combined cohort. Estimates from the Bayesian model closely matched the weighted NHATS estimates. The ability to combine cohorts while generating population-based estimates will allow investigators to address questions that require larger sample sizes, thereby enhancing the value of NHATS to the scientific community.
- 22 September 2025
- Published by Oxford University Press (OUP) in American Journal of Epidemiology
- Vol. 195 (3) , 816-823
Obtaining population-based estimates for survey data using Bayesian hierarchical models with poststratification
For large-scale surveys such as the National Health and Aging Trends Study (NHATS), investigators may wish to combine data from two (or more) cohorts in a single analysis to obtain larger sample sizes. Unfortunately, it is not possible to combine the 2011 and 2015 NHATS cohorts while retaining the sample weights. We applied Bayesian hierarchical models with poststratification as an alternative strategy for obtaining population-based estimates from NHATS. As proof of principle, we compared prevalence estimates of frailty obtained from our Bayesian approach with those obtained from the 2011 and 2015 cohorts using the NHATS sample weights. Once validated, we applied our strategy to combine the cohorts into a single analytical dataset without overlap of participants, and generated Bayesian estimates of frailty for the combined cohort. Estimates from the Bayesian model closely matched the weighted NHATS estimates. The ability to combine cohorts while generating population-based estimates will allow investigators to address questions that require larger sample sizes, thereby enhancing the value of NHATS to the scientific community.
- 7 June 2024
- Published by Oxford University Press (OUP) in The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences
- Vol. 79 (8)
Association Between Cognitive Trajectories and Subsequent Health Status, Depressive Symptoms, and Mortality Among Older Adults in the United States: Findings From a Nationally Representative Study
Background: Cognitive decline may be an early indicator of major health issues in older adults, though research using population-based data is lacking. Researchers objective was to assess the relationships between distinct cognitive trajectories and subsequent health outcomes, including health status, depressive symptoms, and mortality, using a nationally representative cohort. Methods: Data were drawn from the National Health and Aging Trends Study. Global cognition was assessed annually between 2011 and 2018. The health status of 4 413 people, depressive symptoms in 4 342 individuals, and deaths among 5 955 living respondents were measured in 2019. Distinct cognitive trajectory groups were identified using an innovative Bayesian group-based trajectory model. Ordinal logistic, Poisson, and logistic regression models were used to examine the associations between cognitive trajectories and subsequent health outcomes. Results: Researchers identified five cognitive trajectory groups with distinct baseline values and subsequent changes in cognitive function. Compared with the group with stably high cognitive function, worse cognitive trajectories (ie, lower baseline values and sharper declines) were associated with higher risks of poor health status, depressive symptoms, and mortality, even after adjusting for relevant covariates. Conclusions: Among older adults, worse cognitive trajectories are strongly associated with subsequent poor health status, high depressive symptoms, and high mortality risks. Regular screening of cognitive function may help to facilitate early identification and interventions for older adults susceptible to adverse health outcomes.
- 7 June 2024
- Published by Oxford University Press (OUP) in The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences
- Vol. 79 (8)
Association Between Cognitive Trajectories and Subsequent Health Status, Depressive Symptoms, and Mortality Among Older Adults in the United States: Findings From a Nationally Representative Study
Background: Cognitive decline may be an early indicator of major health issues in older adults, though research using population-based data is lacking. Researchers objective was to assess the relationships between distinct cognitive trajectories and subsequent health outcomes, including health status, depressive symptoms, and mortality, using a nationally representative cohort. Methods: Data were drawn from the National Health and Aging Trends Study. Global cognition was assessed annually between 2011 and 2018. The health status of 4 413 people, depressive symptoms in 4 342 individuals, and deaths among 5 955 living respondents were measured in 2019. Distinct cognitive trajectory groups were identified using an innovative Bayesian group-based trajectory model. Ordinal logistic, Poisson, and logistic regression models were used to examine the associations between cognitive trajectories and subsequent health outcomes. Results: Researchers identified five cognitive trajectory groups with distinct baseline values and subsequent changes in cognitive function. Compared with the group with stably high cognitive function, worse cognitive trajectories (ie, lower baseline values and sharper declines) were associated with higher risks of poor health status, depressive symptoms, and mortality, even after adjusting for relevant covariates. Conclusions: Among older adults, worse cognitive trajectories are strongly associated with subsequent poor health status, high depressive symptoms, and high mortality risks. Regular screening of cognitive function may help to facilitate early identification and interventions for older adults susceptible to adverse health outcomes.
- 22 September 2023
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 24 (6)
Benchmarking of local genetic correlation estimation methods using summary statistics from genome-wide association studies
Local genetic correlation evaluates the correlation of additive genetic effects between different traits across the same genetic variants at a genomic locus. It has been proven informative for understanding the genetic similarities of complex traits beyond that captured by global genetic correlation calculated across the whole genome. Several summary-statistics-based approaches have been developed for estimating local genetic correlation, including $\rho$-hess, SUPERGNOVA and LAVA. However, there has not been a comprehensive evaluation of these methods to offer practical guidelines on the choices of these methods. In this study, we conduct benchmark comparisons of the performance of these three methods through extensive simulation and real data analyses. We focus on two technical difficulties in estimating local genetic correlation: sample overlaps across traits and local linkage disequilibrium (LD) estimates when only the external reference panels are available. Our simulations suggest the likelihood of incorrectly identifying correlated regions and local correlation estimation accuracy are highly dependent on the estimation of the local LD matrix. These observations are corroborated by real data analyses of 31 complex traits. Overall, our findings illuminate the distinct results yielded by different methods applied in post-genome-wide association studies (post-GWAS) local correlation studies. We underscore the sensitivity of local genetic correlation estimates and inferences to the precision of local LD estimation. These observations accentuate the vital need for ongoing refinement in methodologies.
- 22 September 2023
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 24 (6)
Benchmarking of local genetic correlation estimation methods using summary statistics from genome-wide association studies
Local genetic correlation evaluates the correlation of additive genetic effects between different traits across the same genetic variants at a genomic locus. It has been proven informative for understanding the genetic similarities of complex traits beyond that captured by global genetic correlation calculated across the whole genome. Several summary-statistics-based approaches have been developed for estimating local genetic correlation, including $\rho$-hess, SUPERGNOVA and LAVA. However, there has not been a comprehensive evaluation of these methods to offer practical guidelines on the choices of these methods. In this study, we conduct benchmark comparisons of the performance of these three methods through extensive simulation and real data analyses. We focus on two technical difficulties in estimating local genetic correlation: sample overlaps across traits and local linkage disequilibrium (LD) estimates when only the external reference panels are available. Our simulations suggest the likelihood of incorrectly identifying correlated regions and local correlation estimation accuracy are highly dependent on the estimation of the local LD matrix. These observations are corroborated by real data analyses of 31 complex traits. Overall, our findings illuminate the distinct results yielded by different methods applied in post-genome-wide association studies (post-GWAS) local correlation studies. We underscore the sensitivity of local genetic correlation estimates and inferences to the precision of local LD estimation. These observations accentuate the vital need for ongoing refinement in methodologies.
- Open Access
Benchmarking local genetic correlation estimation methods using summary statistics from genome-wide association studies
Local genetic correlation evaluates the correlation of genetic effects between different traits across genetic variants in a local region. It has been proven informative for understanding the genetic similarities of complex traits beyond that captured by global genetic correlation calculated across the whole genome. Several summary-statistics-based approaches have been developed for estimating local genetic correlation, including ρ -hess, SUPERGNOVA, and LAVA. However, there has not been a comprehensive evaluation of these methods to offer practical guidelines on the choices of these methods. In this study, we conduct benchmark comparisons of the performance of these three methods through extensive simulation and real data analyses. We focus on two technical difficulties in estimating local genetic correlation: sample overlaps across traits and local linkage disequilibrium (LD) estimates when only the external reference panels are available. Our simulations suggest that the type-I error and estimation accuracy are highly dependent on the estimation of the local LD matrix. These observations are corroborated by real data analyses of 31 complex traits. Overall, our results offer insights into post-GWAS local correlation studies and highlight issues that demand future methodology developments.
Open AccessBenchmarking local genetic correlation estimation methods using summary statistics from genome-wide association studies
Local genetic correlation evaluates the correlation of genetic effects between different traits across genetic variants in a local region. It has been proven informative for understanding the genetic similarities of complex traits beyond that captured by global genetic correlation calculated across the whole genome. Several summary-statistics-based approaches have been developed for estimating local genetic correlation, including ρ -hess, SUPERGNOVA, and LAVA. However, there has not been a comprehensive evaluation of these methods to offer practical guidelines on the choices of these methods. In this study, we conduct benchmark comparisons of the performance of these three methods through extensive simulation and real data analyses. We focus on two technical difficulties in estimating local genetic correlation: sample overlaps across traits and local linkage disequilibrium (LD) estimates when only the external reference panels are available. Our simulations suggest that the type-I error and estimation accuracy are highly dependent on the estimation of the local LD matrix. These observations are corroborated by real data analyses of 31 complex traits. Overall, our results offer insights into post-GWAS local correlation studies and highlight issues that demand future methodology developments.
- Open Access
- 29 August 2022
- Published by MDPI AG in International Journal of Neonatal Screening
- Vol. 8 (3) , 48
dbRUSP: An Interactive Database to Investigate Inborn Metabolic Differences for Improved Genetic Disease Screening
The Recommended Uniform Screening Panel (RUSP) contains more than forty metabolic disorders recommended for inclusion in universal newborn screening (NBS). Tandem-mass-spectrometry-based screening of metabolic analytes in dried blood spot samples identifies most affected newborns, along with a number of false positive results. Due to their influence on blood metabolite levels, continuous and categorical covariates such as gestational age, birth weight, age at blood collection, sex, parent-reported ethnicity, and parenteral nutrition status have been shown to reduce the accuracy of screening. Here, we developed a database and web-based tools (dbRUSP) for the analysis of 41 NBS metabolites and six variables for a cohort of 500,539 screen-negative newborns reported by the California NBS program. The interactive database, built using the R shiny package, contains separate modules to study the influence of single variables and joint effects of multiple variables on metabolite levels. Users can input an individual’s variables to obtain metabolite level reference ranges and utilize dbRUSP to select new candidate markers for the detection of metabolic conditions. The open-source format facilitates the development of data mining algorithms that incorporate the influence of covariates on metabolism to increase accuracy in genetic disease screening.
- 29 August 2022
- Published by MDPI AG in International Journal of Neonatal Screening
- Vol. 8 (3) , 48
Open AccessdbRUSP: An Interactive Database to Investigate Inborn Metabolic Differences for Improved Genetic Disease Screening
The Recommended Uniform Screening Panel (RUSP) contains more than forty metabolic disorders recommended for inclusion in universal newborn screening (NBS). Tandem-mass-spectrometry-based screening of metabolic analytes in dried blood spot samples identifies most affected newborns, along with a number of false positive results. Due to their influence on blood metabolite levels, continuous and categorical covariates such as gestational age, birth weight, age at blood collection, sex, parent-reported ethnicity, and parenteral nutrition status have been shown to reduce the accuracy of screening. Here, we developed a database and web-based tools (dbRUSP) for the analysis of 41 NBS metabolites and six variables for a cohort of 500,539 screen-negative newborns reported by the California NBS program. The interactive database, built using the R shiny package, contains separate modules to study the influence of single variables and joint effects of multiple variables on metabolite levels. Users can input an individual’s variables to obtain metabolite level reference ranges and utilize dbRUSP to select new candidate markers for the detection of metabolic conditions. The open-source format facilitates the development of data mining algorithms that incorporate the influence of covariates on metabolism to increase accuracy in genetic disease screening.