Semi-continuous data present challenges in both model fitting and interpretation. Parametric distributions may be inappropriate for extreme long right tails of the data. Mean effects of covariates, susceptible to extreme values, may fail to capture relevant information for most of the sample. We propose a two-component semi-parametric Bayesian mixture model, with the discrete component captured by a probability mass (typically at zero) and the continuous component of the density modeled by a mixture of B-spline densities that can be flexibly fit to any data distribution. The model includes random effects of subjects to allow for application to longitudinal data. We specify prior distributions on parameters and perform model inference using a Markov chain Monte Carlo (MCMC) Gibbs-sampling algorithm programmed in R. Statistical inference can be made for multiple quantiles of the covariate effects simultaneously providing a comprehensive view. Various MCMC sampling techniques are used to facilitate convergence. We demonstrate the performance and the interpretability of the model via simulations and analyses on the National Consortium on Alcohol and Neurodevelopment in Adolescence study (NCANDA) data on alcohol binge drinking.

With the availability of high-throughput sequencing data, identification of genetic causal variants accurately requires the efficient incorporation of function annotation data into the optimization routine. This motivates the need for development of novel methods for genome wide association studies with special focus on fine-mapping capabilities. A penalty function method that is simple to implement and capable of integrating functional annotation information into the estimation procedure, is proposed in this work. The idea is to use the prior distribution of the effect sizes explicitly as a penalty function. The estimates obtained are shown to be better correlated with the true effect sizes (in comparison with a few existing techniques). An increase in the positive and negative predictive value is demonstrated using Hapgen2 simulated data.

Using individuals' genetic data researchers can generate Polygenic Scores (PS) that are able to predict risk for diseases, variability in different behaviors as well as anthropomorphic measures. This is achieved by leveraging models learned from previously published large Genome-Wide Association Studies (GWASs) associating locations in the genome with a phenotype of interest. Previous GWASs have predominantly been performed in European ancestry individuals. This is of concern as PS generated in samples with a different ancestry to the original training GWAS have been shown to have lower performance and limited portability, and many efforts are now underway to collect genetic databases on individuals of diverse ancestries. In this study, we compare multiple methods of generating PS, including pruning and thresholding and Bayesian continuous shrinkage models, to determine which of them is best able to overcome these limitations. To do this we use the ABCD Study, a longitudinal cohort with deep phenotyping on individuals of diverse ancestry. We generate PS for anthropometric and psychiatric phenotypes using previously published GWAS summary statistics and examine their performance in three subsamples of ABCD: African ancestry individuals (n = 811), European ancestry Individuals (n = 6703), and admixed ancestry individuals (n = 3664). We find that the single ancestry continuous shrinkage method, PRScs (CS), and the multi ancestry meta method, PRScsx Meta (CSx Meta), show the best performance across ancestries and phenotypes.

In the clinical diagnosis of dementia with Lewy bodies, distinction from Alzheimer's disease is suboptimal and complicated by shared genetic risk factors and frequent co-pathology. In the present study we tested the ability of polygenic scores for Alzheimer's disease, dementia with Lewy bodies, and Parkinson's disease to differentiate individuals in a 2713-participant, pathologically defined sample. A dementia with Lewy bodies polygenic score that excluded apolipoprotein E due to its overlap with Alzheimer's disease risk was specifically associated with at least limbic (transitional) Lewy-related pathology and a pathological diagnosis of dementia with Lewy bodies. An Alzheimer's disease polygenic score was associated with neuritic plaques and neurofibrillary tangles but not Lewy-related pathology, and was most strongly associated with an Alzheimer's pathological diagnosis. Our results indicate that an assessment of genetic risk may be useful to clinically distinguish between Alzheimer's disease and dementia with Lewy bodies. Notably, we found no association with a Parkinson's disease polygenic score, which aligns with evidence that dementia with Lewy bodies has a distinct genetic signature that can be exploited to improve clinical diagnoses.

There is a dearth of statistical models that adequately capture the total signal attributed to whole-brain imaging features. The total signal is often widely distributed across the brain, with individual imaging features exhibiting small effect sizes for predicting neurobehavioral phenotypes. The challenge of capturing the total signal is compounded by the distribution of neurobehavioral data, particularly responses to psychological questionnaires, which often feature zero-inflated, highly skewed outcomes. To close this gap, we have developed a novel Variational Bayes algorithm that characterizes the total signal captured by whole-brain imaging features for zero-inflated outcomes. Our zero-inflated variance (ZIV) estimator estimates the fraction of variance explained (FVE) and the proportion of non-null effects (PNN) from large-scale imaging data. In simulations, ZIV demonstrates superior performance over other linear models. When applied to data from the Adolescent Brain Cognitive Development^SM (ABCD) Study, we found that whole-brain imaging features contribute to a larger FVE for externalizing behaviors compared to internalizing behaviors. Moreover, focusing on features contributing to the PNN, ZIV estimator localized key neurocircuitry associated with neurobehavioral traits. To the best of our knowledge, the ZIV estimator is the first specialized method for analyzing zero-inflated neuroimaging data, enhancing future studies on brain-behavior relationships and improving the understanding of neurobehavioral disorders.

Findings in adults have shown that crystallized measures of intelligence, which are more culturally sensitive than fluid intelligence measures, have greater heritability; however, these results have not been found in children. The present study used data from 8,518 participants between 9 and 11 years old from the Adolescent Brain Cognitive Development (ABCD) Study. We found that polygenic predictors of intelligence test performance (based on genome-wide association meta-analyses of data from 269,867 individuals) and of educational attainment (based on data from 1.1 million individuals) predicted neurocognitive performance. We found that crystallized measures were more strongly associated with both polygenic predictors than were fluid measures. This mirrored heritability differences reported previously in adults and suggests similar associations in children. This may be consistent with a prominent role of gene-environment correlation in cognitive development measured by crystallized intelligence tests. Environmental and experiential mediators may represent malleable targets for improving cognitive outcomes.

Background

Genetic risk scores (GRSs) have been associated with CHD events and coronary artery calcium (CAC). We sought to evaluate the ability of a GRS to improve CAC as a screening test.

Methods

Using the results of the most recent genome-wide association studies, we calculated a GRS in 6660 individuals from the Multi-Ethnic Study of Atherosclerosis and used it to determine the optimal age for an individual to undergo CAC screening.

Results

This 157-SNP GRS was predictive of non-zero CAC in individuals aged 44-54 and improved the positive yield of CAC as a screening test in this age group. The GRS was predictive of CAC in the entire multi-ethnic cohort and in each self-identified ethnic group (European American, Chinese American, African American, and Hispanic American) assessed individually. Given a specified target yield rate of non-zero CAC, an equation was derived to calculate an individual's optimal age to undergo CAC screening. In addition, a "direct-to-consumer" GRS consisting of only risk SNPs or their proxies that are directly genotyped on the 23andMe v5 chip (102-SNP GRS) was assessed in the European American population and was predictive of non-zero CAC in younger individuals.

Conclusion

A GRS is associated with non-zero CAC in a multi-ethnic cohort and can be used to calculate the age of a person's first calcium scan, given a target threshold for CAC discovery. Furthermore, an inexpensive and widely available "direct-to-consumer" GRS was found to be a viable option to calculate the optimal age for CAC screening.

Background

Genetic risk scores (GRSs) have been associated with CHD events and coronary artery calcium (CAC). We sought to evaluate the ability of a GRS to improve CAC as a screening test.

Methods

Using the results of the most recent genome-wide association studies, we calculated a GRS in 6660 individuals from the Multi-Ethnic Study of Atherosclerosis and used it to determine the optimal age for an individual to undergo CAC screening.

Results

This 157-SNP GRS was predictive of non-zero CAC in individuals aged 44-54 and improved the positive yield of CAC as a screening test in this age group. The GRS was predictive of CAC in the entire multi-ethnic cohort and in each self-identified ethnic group (European American, Chinese American, African American, and Hispanic American) assessed individually. Given a specified target yield rate of non-zero CAC, an equation was derived to calculate an individual's optimal age to undergo CAC screening. In addition, a "direct-to-consumer" GRS consisting of only risk SNPs or their proxies that are directly genotyped on the 23andMe v5 chip (102-SNP GRS) was assessed in the European American population and was predictive of non-zero CAC in younger individuals.

Conclusion

A GRS is associated with non-zero CAC in a multi-ethnic cohort and can be used to calculate the age of a person's first calcium scan, given a target threshold for CAC discovery. Furthermore, an inexpensive and widely available "direct-to-consumer" GRS was found to be a viable option to calculate the optimal age for CAC screening.