Finding factors influencing risk: comparing Bayesian stochastic search and standard variable selection methods applied to logistic regression models of cases and controls

Adolescent predictors of young adult pain and health outcomes: results from a 6-year prospective follow-up study

ABSTRACT

Adolescent chronic pain may lead to persistent disability and long-term health impairments in adulthood. However, our understanding of which youth are more likely to experience adverse outcomes remains limited. To address this gap, this longitudinal cohort study examined adolescent predictors of various dimensions of young adult health and functioning, including pain, physical health, depression, anxiety, social isolation, and sleep disturbance. As part of a previous clinical trial, we recruited a cohort of adolescents (ages 11-17 years, M age = 14 years) with non-disease-related chronic pain from 15 tertiary pain clinics in North America. Approximately 6 years later, 229 of the original 273 individuals (81% participation rate) completed a follow-up survey as young adults (ages 18-25 years, M age = 21 years). At the young adult follow-up, 73% reported continued chronic pain, with two-thirds experiencing moderate-to-severe pain interference. Youth reported several adverse health outcomes, including below-average physical health (37%), clinically elevated depression (42%), clinically elevated anxiety (48%), and sleep disturbances (77%). Multivariate regression analyses controlling for sociodemographic characteristics revealed that higher pain intensity, more pain locations, lower sleep quality, and greater anxiety symptoms in adolescence predicted worse pain outcomes in young adulthood. Moreover, lower sleep quality, greater anxiety symptoms, and worse family functioning predicted worse physical and psychosocial health in adulthood. These findings represent an important first step toward identifying ways to optimize psychological pain interventions. Tailored psychological pain interventions can directly target adolescent vulnerabilities, including mood, sleep, and family risk factors, with the potential to disrupt a lifelong trajectory of pain and suffering.

Neural and psychological correlates of post-traumatic stress symptoms in a community adult sample

A multitude of factors are associated with the symptoms of post-traumatic stress disorder. However, establishing which predictors are most strongly associated with post-traumatic stress disorder symptoms is complicated because few studies are able to consider multiple factors simultaneously across the biopsychosocial domains that are implicated by existing theoretical models. Further, post-traumatic stress disorder is heterogeneous, and studies using case-control designs may obscure which factors relate uniquely to symptom dimensions. Here we used Bayesian variable selection to identify the most important predictors for overall post-traumatic stress disorder symptoms and individual symptom dimensions in a community sample of 569 adults (18 to 85 yr of age). Candidate predictors were selected from previously established risk factors relevant for post-traumatic stress disorder and included psychological measures, behavioral measures, and resting state functional connectivity among brain regions. In a follow-up analysis, we compared results controlling for current depression symptoms in order to examine specificity. Poor sleep quality and dimensions of temperament and impulsivity were consistently associated with greater post-traumatic stress disorder symptom severity. In addition to self-report measures, brain functional connectivity among regions commonly ascribed to the default mode network, central executive network, and salience network explained the unique variability of post-traumatic stress disorder symptoms. This study demonstrates the unique contributions of psychological measures and neural substrates to post-traumatic stress disorder symptoms.

Added value of arterial enhancement fraction derived from dual-energy computed tomography for preoperative diagnosis of cervical lymph node metastasis in papillary thyroid cancer: initial results

OBJECTIVES

To explore the added value of arterial enhancement fraction (AEF) derived from dual-energy computed tomography CT (DECT) to conventional image features for diagnosing cervical lymph node (LN) metastasis in papillary thyroid cancer (PTC).

METHODS

A total of 273 cervical LNs (153 non-metastatic and 120 metastatic) were recruited from 92 patients with PTC. Qualitative image features of LNs were assessed. Both single-energy CT (SECT)-derived AEF (AEFS) and DECT-derived AEF (AEFD) were calculated. Correlation between AEFD and AEFS was determined using Pearson's correlation coefficient. Multivariate logistic regression analysis with the forward variable selection method was used to build three models (conventional features, conventional features + AEFS, and conventional features + AEFD). Diagnostic performances were evaluated using receiver operating characteristic (ROC) curve analyses.

RESULTS

Abnormal enhancement, calcification, and cystic change were chosen to build model 1 and the model provided moderate diagnostic performance with an area under the ROC curve (AUC) of 0.675. Metastatic LNs demonstrated both significantly higher AEFD (1.14 vs 0.48; p < 0.001) and AEFS (1.08 vs 0.38; p < 0.001) than non-metastatic LNs. AEFD correlated well with AEFS (r = 0.802; p < 0.001), and exhibited comparable performance with AEFS (AUC, 0.867 vs 0.852; p = 0.628). Combining CT image features with AEFS (model 2) and AEFD (model 3) could significantly improve diagnostic performances (AUC, 0.865 vs 0.675; AUC, 0.883 vs 0.675; both p < 0.001).

CONCLUSIONS

AEFD correlated well with AEFS, and exhibited comparable performance with AEFS. Integrating qualitative CT image features with both AEFS and AEFD could further improve the ability in diagnosing cervical LN metastasis in PTC.

CLINICAL RELEVANCE STATEMENT

Arterial enhancement fraction (AEF) values, especially AEF derived from dual-energy computed tomography, can help to diagnose cervical lymph node metastasis in patients with papillary thyroid cancer, and complement conventional CT image features for improved clinical decision making.

KEY POINTS

• Metastatic cervical lymph nodes (LNs) demonstrated significantly higher arterial enhancement fraction (AEF) derived from dual-energy computed tomography (DECT) and single-energy CT (SECT)-derived AEF (AEFS) than non-metastatic LNs in patients with papillary thyroid cancer. • DECT-derived AEF (AEFD) correlated significantly with AEFS, and exhibited comparable performance with AEFS. • Integrating qualitative CT images features with both AEFS and AEFD could further improve the differential ability.

Factors Associated with Antiretroviral Therapy Adherence Among a Community-Based Sample of Sexual Minority Older Adults with HIV

Older sexual minorities (e.g., gay, bisexual) living with HIV are at risk for poor HIV outcomes due to their frequent experience with both psychosocial challenges and structural barriers to care. This study utilized a stochastic search variable selection (SVSS) approach to explore potential psychosocial and structural factors associated with HIV-related health outcomes among a community-based sample of older sexual minorities (N = 150) in South Florida, an U.S. HIV-epidemic epicenter. After SVSS, a forward entry regression approach suggested unstable housing, illicit substance use, current nicotine use, and depression were all associated with poorer ART adherence among older sexual minority adults living with HIV. No associations between potential correlates and biological measures of HIV disease severity were observed. Findings highlight a need to focus on multiple levels of intervention that target a combination of psychosocial and structural factors to improve HIV-care outcomes among older sexual minorities and achieve Ending the HIV Epidemic goals.

Comparing Bayesian Variable Selection to Lasso Approaches for Applications in Psychology

In the current paper, we review existing tools for solving variable selection problems in psychology. Modern regularization methods such as lasso regression have recently been introduced in the field and are incorporated into popular methodologies, such as network analysis. However, several recognized limitations of lasso regularization may limit its suitability for psychological research. In this paper, we compare the properties of lasso approaches used for variable selection to Bayesian variable selection approaches. In particular we highlight advantages of stochastic search variable selection (SSVS), that make it well suited for variable selection applications in psychology. We demonstrate these advantages and contrast SSVS with lasso type penalization in an application to predict depression symptoms in a large sample and an accompanying simulation study. We investigate the effects of sample size, effect size, and patterns of correlation among predictors on rates of correct and false inclusion and bias in the estimates. SSVS as investigated here is reasonably computationally efficient and powerful to detect moderate effects in small sample sizes (or small effects in moderate sample sizes), while protecting against false inclusion and without over-penalizing true effects. We recommend SSVS as a flexible framework that is well-suited for the field, discuss limitations, and suggest directions for future development.

Factors associated with early childhood development in municipalities of Ceará, Brazil: a hierarchical model of contexts, environments, and nurturing care domains in a cross-sectional study

Background

This study aims to identify the contexts, environments, and nurturing care predictors that determine whether a child is developmentally on track in Ceará, Brazil.

Methods

We analysed data from a cross-sectional study conducted with caregivers of 6,447 children aged 0-59 months during a vaccination campaign in Ceará in October 2019. The validated Child Development Assessment Questionnaire was used to assess early childhood development (ECD) and children with a z-score ≥ -1 SD were considered developmentally on track. We conducted logistic regression models to understand the effects of contexts, environments, and nurturing care domains on ECD.

Findings

Children in the early years (< 36 months) were more likely to meet the ECD milestones if they were not born with low birth weight (AOR: 0·64; 95% CI: 0·42-0·97), were exposed to manufactured toys in their house (2·68; 1·97-3·66), their heads of household were employed (1·61; 1·16-2·23), and their caregivers had read the Child Health Handbook (1·42; 1·13-1·77) and engaged them in stimulating activities (1·71; 1·26-2·32). Children aged 36-59 months were more likely to meet the ECD milestones if they were breastfed (never: ref. / < 3 months: 3·72; 1·91-7·26 / 3-5 months: 3·21; 1·74-5·93 / 6-11 months: 3·73; 1·95-7·16 / ≥ 12 months: 3·89; 2·25-6·72), had books at home (0: ref / 1-3: 1·71; 1·22-2·40 / 4-6: 2·24; 1·27-3·94 / 7+: 2·71; 1·05-7·00), and their caregivers received information about ECD (1·49; 1·11-2·01) and engaged them in stimulating activities (1·80; 1·27-2·56). Children aged 36-59 months were less likely to meet developmental milestones if they watched TV or used tablets/smartphones for more than two hours per day (0·61; 0·44-0·84), played with household objects (0·62; 0·41-0·92), participated in governmental early childhood programmes aimed at vulnerable families (0·62; 0·45-0·86), had families that participated in income transfer programmes (0·68; 0·47-0·99) (families living in poverty or extreme poverty), and their caregivers considered slapping (0·67; 0·48-0·94) a necessary disciplinary method.

Interpretation

Having favourable socioeconomic conditions, breastfeeding, the absence of harsh discipline, caregivers who provide responsive care, and the provision of opportunities for early learning are the key factors that increase the likelihood of a child achieving their full developmental potential in Ceará, Brazil.

Funding

This study was supported by the Maria Cecília Souto Vidigal Foundation (F0245), Brazil. The funder had no role in the design, analysis, or writing of this article.

Patterns and Drivers of Household Sanitation Access and Sustainability in Kwale County, Kenya

Many sanitation interventions suffer from poor sustainability. Failure to maintain or replace toilet facilities risks exposing communities to environmental pathogens, yet little is known about the factors that drive sustained access beyond project life spans. Using data from a cohort of 1666 households in Kwale County, Kenya, we investigated the factors associated with changes in sanitation access between 2015 and 2017. Sanitation access is defined as access to an improved or unimproved facility within the household compound that is functional and in use. A range of contextual, psychosocial, and technological covariates were included in logistic regression models to estimate their associations with (1) the odds of sustaining sanitation access and (2) the odds of gaining sanitation access. Over two years, 28.3% households sustained sanitation access, 4.7% lost access, 17.7% gained access, and 49.2% remained without access. Factors associated with increased odds of households sustaining sanitation access included not sharing the facility and presence of a solid washable slab. Factors associated with increased odds of households gaining sanitation access included a head with at least secondary school education, level of coarse soil fragments, and higher local sanitation coverage. Results from this study can be used by sanitation programs to improve the rates of initial and sustained adoption of sanitation.

Syndemics and salivary inflammation in people living with HIV/AIDS

OBJECTIVE

People living with HIV/AIDS (PLWHA) are disproportionally exposed to a host of structural, community, and individual-level physical and psychosocial stressors also termed 'syndemic conditions.' The current study aimed to examine the association between experiencing syndemic conditions and physiological stress response and be associated with bodily inflammation, including Interlekin-6 (IL-6) and C-reactive protein (CRP) in PLWHA.

DESIGN

Participants (N = 103) were recruited from a public HIV clinic. They provided saliva samples of IL-6 and CRP and completed psychosocial measures.

MAIN OUTCOME MEASURES

Levels of circulating salivary IL-6 and CRP.

RESULTS

When predictors (birth country, recent housing instability, and incarceration history) were simultaneously entered into a regression model, only incarceration history was negatively associated with IL-6 [b = -.27, t(98) = -3.11, p = .002]. For CRP, the resulting regression model was not significant, [F(3, 98) = 2.23, p = .090].

CONCLUSION

Although we had expected higher levels of syndemics to be associated with higher levels of circulating inflammation, in our sample, length of incarceration was associated with lower levels of circulating IL-6. Findings are therefore suggestive of a stress response disruption resulting in a negative feedback loop as the long-term impact of chronic stress on inflammation.

The role of an immune signature for prognosis and immunotherapy response in endometrial cancer

Immunotherapy is a practical and promising treatment for advanced and recurrent endometrial cancer (EC). In this study, we identified an immune-related gene (IRG) signature to predict the overall survival (OS) and response to immune checkpoints inhibitors (ICIs) in patients with EC. The RNA expression profiles of EC were obtained from The Cancer Genome Atlas database and then were filtered for IRGs based on the Immport database. Using the conjoint Cox regression model, an immune signature consisting of seven risk IRGs (CBLC, PLA2G2A, TNF, NR3C1, APOD, TNFRSF18, and LTB) was developed. The immune signature was independent of other clinical factors and was superior to the traditional staging method for OS prediction in EC. Immunohistochemistry staining from the Human Protein Atlas database and quantitative real-time PCR analysis of EC samples were also performed to validate the expression levels of risk IRGs. By further analyzing the tumor microenvironment in EC, patients in the low-risk subgroup showed a higher immune cell infiltration status, which was associated with a better prognosis. Moreover, the tumor mutational burden and immunophenoscore analysis demonstrated that the low-risk subgroup was more sensitive to ICI-based immunotherapy. These findings might shed light on the development of targeted treatment and novel biomarkers for patients with EC.

Performance of variable and function selection methods for estimating the nonlinear health effects of correlated chemical mixtures: A simulation study

Statistical methods for identifying harmful chemicals in a correlated mixture often assume linearity in exposure-response relationships. Nonmonotonic relationships are increasingly recognized (eg, for endocrine-disrupting chemicals); however, the impact of nonmonotonicity on exposure selection has not been evaluated. In a simulation study, we assessed the performance of Bayesian kernel machine regression (BKMR), Bayesian additive regression trees (BART), Bayesian structured additive regression with spike-slab priors (BSTARSS), generalized additive models with double penalty (GAMDP) and thin plate shrinkage smoothers (GAMTS), multivariate adaptive regression splines (MARS), and lasso penalized regression. We simulated realistic exposure data based on pregnancy exposure to 17 phthalates and phenols in the US National Health and Nutrition Examination Survey using a multivariate copula. We simulated data sets of size N = 250 and compared methods across 32 scenarios, varying by model size and sparsity, signal-to-noise ratio, correlation structure, and exposure-response relationship shapes. We compared methods in terms of their sensitivity, specificity, and estimation accuracy. In most scenarios, BKMR, BSTARSS, GAMDP, and GAMTS achieved moderate to high sensitivity (0.52-0.98) and specificity (0.21-0.99). BART and MARS achieved high specificity (≥0.90), but low sensitivity in low signal-to-noise ratio scenarios (0.20-0.51). Lasso was highly sensitive (0.71-0.99), except for quadratic relationships (≤0.27). Penalized regression methods that assume linearity, such as lasso, may not be suitable for studies of environmental chemicals hypothesized to have nonmonotonic relationships with outcomes. Instead, BKMR, BSTARSS, GAMDP, and GAMTS are attractive methods for flexibly estimating the shapes of exposure-response relationships and selecting among correlated exposures.

Improving Practices for Selecting a Subset of Important Predictors in Psychology: An Application to Predicting Pain

Frequently, researchers in psychology are faced with the challenge of narrowing down a large set of predictors to a smaller subset. There are a variety of ways to do this, but commonly it is done by choosing predictors with the strongest bivariate correlations with the outcome. However, when predictors are correlated, bivariate relationships may not translate into multivariate relationships. Further, any attempts to control for multiple testing are likely to result in extremely low power. Here we introduce a Bayesian variable-selection procedure frequently used in other disciplines, stochastic search variable selection (SSVS). We apply this technique to choosing the best set of predictors of the perceived unpleasantness of an experimental pain stimulus from among a large group of sociocultural, psychological, and neurobiological (functional MRI) individual-difference measures. Using SSVS provides information about which variables predict the outcome, controlling for uncertainty in the other variables of the model. This approach yields new, useful information to guide the choice of relevant predictors. We have provided Web-based open-source software for performing SSVS and visualizing the results.

Development and Validation of a Novel Evidence-Based Lupus Multivariable Outcome Score for Clinical Trials

OBJECTIVE

Trials of new systemic lupus erythematosus (SLE) treatments are hampered by the lack of effective outcome measures. To address this, we developed a novel Lupus Multivariable Outcome Score (LuMOS) and assessed its performance using data from 2 randomized controlled trials of belimumab in patients with SLE.

METHODS

The LuMOS formula was developed by analyzing raw data from 2 pivotal trials, the Study of Belimumab in Subjects with SLE 52-week (BLISS-52) and 76-week (BLISS-76) trials, which are the basis for approval of belimumab. Using the BLISS-76 trial data as the learning data set, we carried out multivariable logistic regression analyses to optimize discrimination of outcomes between patients treated with 10 mg/kg belimumab and patients receiving placebo over the first 52 weeks of follow-up. In addition, the performance of LuMOS was assessed using an independent validation data set from the BLISS-52 trial.

RESULTS

The LuMOS model incorporated the following response criteria: a ≥4-point reduction on the Safety of Estrogens in Lupus Erythematosus National Assessment version of the SLE Disease Activity Index, an increase in C4 levels, a decrease in anti-double-stranded DNA titers, and changes in the British Isles Lupus Assessment Group scores for organ system manifestations (no worsening in renal components, and improvements in mucocutaneous components). A decrease in the prednisone dose and increase in C3 levels had very minor impacts on the total LuMOS score. In all analyses of the BLISS-76 and BLISS-52 trial data sets, the mean LuMOS scores were significantly higher (P < 0.0001) in patients treated with 1 mg or 10 mg belimumab compared to placebo. In contrast to the performance of the SLE Responder Index 4 (SRI-4), the LuMOS revealed significant differences between the active treatment group (1 mg belimumab in the BLISS-76 cohort) and placebo group. The effect sizes were significantly much higher with the LuMOS than with the SRI-4.

CONCLUSION

The evidenced-based LuMOS outcome scoring system, developed with data from the BLISS-76 trial of belimumab in patients with SLE and validated with data from the BLISS-52 trial, exhibits a superior capacity to discriminate responders from nonresponders when compared to the SRI-4. Use of the LuMOS may improve the efficiency and power of analyses in future lupus trials.

Northern Shanghai Study: cardiovascular risk and its associated factors in the Chinese elderly-a study protocol of a prospective study design

INTRODUCTION

Cardiovascular (CV) diseases are the leading cause of death and disability in the world. Increasing lifespans and ageing populations also contribute to an increasing CV burden. However, in China, there were few well-designed cohort studies focusing on the elderly population, let alone an established CV risk score. The objective of this study is to establish a CV risk score based on a community-dwelling Chinese elderly population, determining the profile of the associated CV risk factors and target organ damages (TODs), so as to guide the later intervention.

METHODS AND ANALYSIS

The Northern Shanghai Study is an ongoing prospective community-based study. After enrolment, clinical examination, anthropometric measurement and a questionnaire will be administered to each participant at baseline and after every 2 years in the follow-up. Our tests and examinations include: blood/urine sample and biochemical measurements, office blood pressure recording, carotid ultrasonograph, echocardiograph, pulse wave velocity, pulse wave analysis, 4-limb blood pressure recording, body mass index, etc. Baseline measurement will also include the assessments on TODs and the conventional CV risk factors. In the follow-up, the incidence of CV events and mortality will be recorded. The Northern Shanghai Risk Score will be calculated, with considerations on CV risk factors and TODs.

ETHICS AND DISSEMINATION

This study was approved by the Shanghai Tenth People's Hospital Institutional Review Board. All participants signed a written consent form.

TRIAL REGISTRATION NUMBER

NCT02368938; Pre-results.

Clinical risk model assessment for cardiovascular autonomic dysfunction in the general Chinese population

BACKGROUND

The aim of this study was to evaluate the prevalence of cardiovascular autonomic (CA) dysfunction in the general Chinese population (instead of focusing on only patients with diabetes) and to develop a clinical risk model for the disease.

METHODS AND MATERIALS

We evaluated CA dysfunction prevalence in a dataset based on a population sample consisting of 2,092 individuals. Clinical risk models were derived from exploratory sets using multiple logistic regression analysis. The performance of the clinical risk models was tested in the validation sets.

RESULTS

CA dysfunction prevalence was 18.50% in the general Chinese population, while the prevalence was 24.14% in individuals aged ≥60 years. Its prevalence was 31.17, 24.69, and 21.26% in patients with diabetes, and hypertensive, and metabolic syndrome populations, respectively. Finally, we developed clinical risk models involving seven risk factors. The mean area under the receiver-operating curve was 0.758 (95% CI 0.724-0.793) for these models. The mean sensitivity and specificity of the clinical risk models was 75.0 and 66.2%, respectively.

CONCLUSION

CA dysfunction prevalence was high in the general Chinese population, and its prevalence was more frequent in individuals with diabetes, and hypertensive, and metabolic syndrome. Clinical risk models with a high value for predicting CA dysfunction were developed. CA dysfunction has become a major public health problem in China that requires strategies aimed at the prevention and treatment of the disease.

Air toxics and birth defects: a Bayesian hierarchical approach to evaluate multiple pollutants and spina bifida

BACKGROUND

While there is evidence that maternal exposure to benzene is associated with spina bifida in offspring, to our knowledge there have been no assessments to evaluate the role of multiple hazardous air pollutants (HAPs) simultaneously on the risk of this relatively common birth defect. In the current study, we evaluated the association between maternal exposure to HAPs identified by the United States Environmental Protection Agency (U.S. EPA) and spina bifida in offspring using hierarchical Bayesian modeling that includes Stochastic Search Variable Selection (SSVS).

METHODS

The Texas Birth Defects Registry provided data on spina bifida cases delivered between 1999 and 2004. The control group was a random sample of unaffected live births, frequency matched to cases on year of birth. Census tract-level estimates of annual HAP levels were obtained from the U.S. EPA's 1999 Assessment System for Population Exposure Nationwide. Using the distribution among controls, exposure was categorized as high exposure (>95(th) percentile), medium exposure (5(th)-95(th) percentile), and low exposure (<5(th) percentile, reference). We used hierarchical Bayesian logistic regression models with SSVS to evaluate the association between HAPs and spina bifida by computing an odds ratio (OR) for each HAP using the posterior mean, and a 95% credible interval (CI) using the 2.5(th) and 97.5(th) quantiles of the posterior samples. Based on previous assessments, any pollutant with a Bayes factor greater than 1 was selected for inclusion in a final model.

RESULTS

Twenty-five HAPs were selected in the final analysis to represent "bins" of highly correlated HAPs (ρ > 0.80). We identified two out of 25 HAPs with a Bayes factor greater than 1: quinoline (ORhigh = 2.06, 95% CI: 1.11-3.87, Bayes factor = 1.01) and trichloroethylene (ORmedium = 2.00, 95% CI: 1.14-3.61, Bayes factor = 3.79).

CONCLUSIONS

Overall there is evidence that quinoline and trichloroethylene may be significant contributors to the risk of spina bifida. Additionally, the use of Bayesian hierarchical models with SSVS is an alternative approach in the evaluation of multiple environmental pollutants on disease risk. This approach can be easily extended to environmental exposures, where novel approaches are needed in the context of multi-pollutant modeling.

A Bayesian approach to identify genes and gene-level SNP aggregates in a genetic analysis of cancer data

Complex diseases, such as cancer, arise from complex etiologies consisting of multiple single-nucleotide polymorphisms (SNPs), each contributing a small amount to the overall risk of disease. Thus, many researchers have gone beyond single-SNPs analysis methods, focusing instead on groups of SNPs, for example by analysing haplotypes. More recently, pathway-based methods have been proposed that use prior biological knowledge on gene function to achieve a more powerful analysis of genome-wide association studies (GWAS) data. In this paper we propose a novel Bayesian modeling framework to identify molecular biomarkers for disease prediction. Our method combines pathway-based approaches with multiple SNP analyses of a specified region of interest. The model's development is motivated by SNP data from a lung cancer study. In our approach we define gene-level scores based on SNP allele frequencies and use a linear modeling setting to study the scores association to the observed phenotype. The basic idea behind the definition of gene-level scores is to weigh the SNPs within the gene according to their rarity, based on genotype frequencies expected under the Hardy-Weinberg equilibrium law. This results in scores giving more importance to the unusually low frequencies, i.e. to SNPs that might indicate peculiar genetic differences between subjects belonging to different groups. An additional feature of our approach is that we incorporate information on SNP-to-SNP associations into the model. In particular, we use network priors that model the linkage disequilibrium between SNPs. For posterior inference, we design a stochastic search method that identifies significant biomarkers (genes and SNPs) for disease prediction. We assess performances on simulated data and compare results to existing approaches. We then show the ability of the proposed methodology to detect relevant genes and associated SNPs in a lung cancer dataset.

Small sample properties of rare variant analysis methods

We are now well into the sequencing era of genetic analysis, and methods to investigate rare variants associated with disease remain in high demand. Currently, the more common rare variant analysis methods are burden tests and variance component tests. This report introduces a burden test known as the modified replication based sum statistic and evaluates its performance, and the performance of other common burden and variance component tests under the setting of a small sample size (103 total cases and controls) using the Genetic Analysis Workshop 18 simulated data with complete knowledge of the simulation model. Specifically we look at the variable threshold sum statistic, replication-based sum statistics, the C-alpha, and sequence kernel association test. Using minor allele frequency thresholds of less than 0.05, we find that the modified replication based sum statistic is competitive with all methods and that using 103 individuals leads to all methods being vastly underpowered. Much larger sample sizes are needed to confidently find truly associated genes.

Bayesian Variable Selection under the Proportional Hazards Mixed-effects Model

Over the past decade much statistical research has been carried out to develop models for correlated survival data; however, methods for model selection are still very limited. A stochastic search variable selection (SSVS) approach under the proportional hazards mixed-effects model (PHMM) is developed. The SSVS method has previously been applied to linear and generalized linear mixed models, and to the proportional hazards model with high dimensional data. Because the method has mainly been developed for hierarchical normal mixture distributions, it operates on the linear predictor under the Cox type models. The PHMM naturally incorporates the normal distribution via the random effects, which enables SSVS to efficiently search through the candidate variable space. The approach was evaluated through simulation, and applied to a multi-center lung cancer clinical trial data set, for which the variable selection problem was previously debated upon in the literature.

A simple Chinese risk score model for screening cardiovascular autonomic neuropathy

Background

The purpose of the present study was to develop and evaluate a risk score to predict people at high risk of cardiovascular autonomic dysfunction neuropathy (CAN) in Chinese population.

Methods and Materials

A population-based sample of 2,092 individuals aged 30–80 years, without previously diagnosed CAN, was surveyed between 2011 and 2012. All participants underwent short-term HRV test. The risk score was derived from an exploratory set. The risk score was developed by stepwise backward multiple logistic regression. The coefficients from this model were transformed into components of a CAN score. This score was tested in a validation and entire sample.

Results

The final risk score included age, body mass index, hypertension, resting hear rate, items independently and significantly (P<0.05) associated with the presence of previously undiagnosed CAN. The area under the receiver operating curve was 0.726 (95% CI 0.686–0.766) for exploratory set, 0.784 (95% CI 0.749–0.818) for validation set, and 0.756 (95% CI 0.729–0.782) for entire sample. In validation set, at optimal cutoff score of 5 of 10, the risk score system has the sensitivity, specificity, and percentage that needed subsequent testing were 69, 78, and 30%, respectively.

Conclusion

We developed a CAN risk score system based on a set of variables not requiring laboratory tests. The score system is simple fast, inexpensive, noninvasive, and reliable tool that can be applied to early intervention to delay or prevent the disease in China.

Using a Bayesian hierarchical model for identifying single nucleotide polymorphisms associated with childhood acute lymphoblastic leukemia risk in case-parent triads

Childhood acute lymphoblastic leukemia (ALL) is a condition that arises from complex etiologies. The absence of consistent environmental risk factors and the presence of modest familial associations suggest ALL is a complex trait with an underlying genetic component. The identification of genetic factors associated with disease is complicated by complex genetic covariance structures and multiple testing issues. Both issues can be resolved with appropriate Bayesian variable selection methods. The present study was undertaken to extend our hierarchical Bayesian model for case-parent triads to incorporate single nucleotide polymorphisms (SNPs) and incorporate the biological grouping of SNPs within genes. Based on previous evidence that genetic variation in the folate metabolic pathway influences ALL risk, we evaluated 128 tagging SNPs in 16 folate metabolic genes among 118 ALL case-parent triads recruited from the Texas Children’s Cancer Center (Houston, TX) between 2003 and 2010. We used stochastic search gene suggestion (SSGS) in hierarchical Bayesian models to evaluate the association between folate metabolic SNPs and ALL. Using Bayes factors among these variants in childhood ALL case-parent triads, two SNPs were identified with a Bayes factor greater than 1. There was evidence that the minor alleles of NOS3 rs3918186 (OR = 2.16; 95% CI: 1.51-3.15) and SLC19A1 rs1051266 (OR = 2.07; 95% CI: 1.25-3.46) were positively associated with childhood ALL. Our findings are suggestive of the role of inherited genetic variation in the folate metabolic pathway on childhood ALL risk, and they also suggest the utility of Bayesian variable selection methods in the context of case-parent triads for evaluating the role of SNPs on disease risk.

Artificial neural network models for prediction of cardiovascular autonomic dysfunction in general Chinese population

BACKGROUND

The present study aimed to develop an artificial neural network (ANN) based prediction model for cardiovascular autonomic (CA) dysfunction in the general population.

METHODS

We analyzed a previous dataset based on a population sample consisted of 2,092 individuals aged 30-80 years. The prediction models were derived from an exploratory set using ANN analysis. Performances of these prediction models were evaluated in the validation set.

RESULTS

Univariate analysis indicated that 14 risk factors showed statistically significant association with CA dysfunction (P < 0.05). The mean area under the receiver-operating curve was 0.762 (95% CI 0.732-0.793) for prediction model developed using ANN analysis. The mean sensitivity, specificity, positive and negative predictive values were similar in the prediction models was 0.751, 0.665, 0.330 and 0.924, respectively. All HL statistics were less than 15.0.

CONCLUSION

ANN is an effective tool for developing prediction models with high value for predicting CA dysfunction among the general population.

Symptom clusters of pain, depressed mood, and fatigue in lung cancer: assessing the role of cytokine genes

PURPOSE

Symptom clusters, the multiple, co-occurring symptoms experienced by cancer patients, are debilitating and affects quality of life. We assessed if a panel of immune-response genes may underlie the co-occurrence of severe pain, depressed mood, and fatigue and help identify patients with severe versus non-severe symptom clusters.

METHODS

Symptoms were assessed at presentation, prior to cancer treatment in 599 newly diagnosed lung cancer patients. We applied cluster analyses to determine the patients with severe versus non-severe symptom clusters of pain, depressed mood, and fatigue.

RESULTS

Two homogenous clusters were identified. One hundred sixteen patients (19 %) comprised the severe symptom cluster, reporting high intensity of pain, depressed mood, and fatigue and 183 (30 %) patients reported low intensity of these symptoms. Using Bayesian model averaging methodology, we found that of the 55 single nucleotide polymorphisms assessed, an additive effect of mutant alleles in endothelial nitric oxide synthase (-1474 T/A) (posterior probability of inclusion (PPI) = 0.78, odds ratio (OR) = 0.54, 95 % credible interval (CI) = (0.31, 0.93)); IL1B T-31C (PPI = 0.72, OR = 0.55, 95 % CI = (0.31, 0.97)); TNFR2 Met(196)Arg (PPI = 0.70, OR = 1.85, 95 % CI = (1.03, 3.36)); PTGS2 exon 10+837T > C (PPI = 0.69, OR = 0.54, 95 % CI = (0.28, 0.99)); and IL10RB Lys(47)Glu (PPI = 0.68; OR = 1.74; 95 % CI = (1.04, 2.92)) were predictive for symptom clusters.

CONCLUSIONS

Genetic polymorphisms may facilitate identification of high-risk patients and development of individualized symptom therapies.

Investigating multiple candidate genes and nutrients in the folate metabolism pathway to detect genetic and nutritional risk factors for lung cancer

PURPOSE

Folate metabolism, with its importance to DNA repair, provides a promising region for genetic investigation of lung cancer risk. This project investigates genes (MTHFR, MTR, MTRR, CBS, SHMT1, TYMS), folate metabolism related nutrients (B vitamins, methionine, choline, and betaine) and their gene-nutrient interactions.

METHODS

We analyzed 115 tag single nucleotide polymorphisms (SNPs) and 15 nutrients from 1239 and 1692 non-Hispanic white, histologically-confirmed lung cancer cases and controls, respectively, using stochastic search variable selection (a Bayesian model averaging approach). Analyses were stratified by current, former, and never smoking status.

RESULTS

Rs6893114 in MTRR (odds ratio [OR] = 2.10; 95% credible interval [CI]: 1.20-3.48) and alcohol (drinkers vs. non-drinkers, OR = 0.48; 95% CI: 0.26-0.84) were associated with lung cancer risk in current smokers. Rs13170530 in MTRR (OR = 1.70; 95% CI: 1.10-2.87) and two SNP*nutrient interactions [betaine*rs2658161 (OR = 0.42; 95% CI: 0.19-0.88) and betaine*rs16948305 (OR = 0.54; 95% CI: 0.30-0.91)] were associated with lung cancer risk in former smokers. SNPs in MTRR (rs13162612; OR = 0.25; 95% CI: 0.11-0.58; rs10512948; OR = 0.61; 95% CI: 0.41-0.90; rs2924471; OR = 3.31; 95% CI: 1.66-6.59), and MTHFR (rs9651118; OR = 0.63; 95% CI: 0.43-0.95) and three SNP*nutrient interactions (choline*rs10475407; OR = 1.62; 95% CI: 1.11-2.42; choline*rs11134290; OR = 0.51; 95% CI: 0.27-0.92; and riboflavin*rs8767412; OR = 0.40; 95% CI: 0.15-0.95) were associated with lung cancer risk in never smokers.

CONCLUSIONS

This study identified possible nutrient and genetic factors related to folate metabolism associated with lung cancer risk, which could potentially lead to nutritional interventions tailored by smoking status to reduce lung cancer risk.

A Bayesian integrative genomic model for pathway analysis of complex traits

With new technologies, multiple types of genomic data are commonly collected on a single set of samples. However, standard analysis methods concentrate on a single data type at a time and ignore the relationships between genes, proteins, and biochemical reactions that give rise to complex phenotypes. In this paper, we propose a novel integrative model to incorporate multiple types of genomic data into an association analysis with a complex phenotype. The method combines path analysis and stochastic search variable selection into a Bayesian hierarchical model that simultaneously identifies both direct and indirect genomic effects on the phenotype. Results from a simulation study and application of the Bayesian model to a pharmacogenomic study of the drug gemcitabine demonstrate greater sensitivity to detect genomic effects in some simulation scenarios, when compared to the standard single data type analysis. Further research is required to extend and modify this integrative modeling framework to increase computational efficiency to best capitalize on the wealth of information available across multiple genomic data types.

Objective Bayes model selection in probit models

We describe a new variable selection procedure for categorical responses where the candidate models are all probit regression models. The procedure uses objective intrinsic priors for the model parameters, which do not depend on tuning parameters, and ranks the models for the different subsets of covariates according to their model posterior probabilities. When the number of covariates is moderate or large, the number of potential models can be very large, and for those cases, we derive a new stochastic search algorithm that explores the potential sets of models driven by their model posterior probabilities. The algorithm allows the user to control the dimension of the candidate models and thus can handle situations when the number of covariates exceed the number of observations. We assess, through simulations, the performance of the procedure and apply the variable selector to a gene expression data set, where the response is whether a patient exhibits pneumonia. Software needed to run the procedures is available in the R package varselectIP.

Using Ascertainment for Targeted Resequencing to Increase Power to Identify Causal Variants

Researchers continue to use genome-wide association studies (GWAS) to find the genetic markers associated with disease. Recent studies have added to the typical two-stage analysis a third stage that uses targeted resequencing on a randomly selected subset of the cases to detect the causal single-nucleotide polymorphism (SNP). We propose a design for targeted resequencing that increases the power to detect the causal variant. The design features an ascertainment scheme wherein only those cases with the presence of a risk allele are selected for targeted resequencing. We simulated a disease with a single causal SNP to evaluate our method versus a targeted resequencing design using randomly selected individuals. The simulation studies showed that ascertaining individuals for the targeted resequencing can substantially increase the power to detect a causal SNP, without increasing the false-positive rate.

Pooling dietary data using questionnaires with open-ended and predefined responses: implications for comparing mean intake or estimating odds ratios

In the current era of diet-gene analyses, large sample sizes are required to uncover the etiology of complex diseases. As such, consortia form and often combine available data. Food frequency questionnaires, which commonly use 2 different types of responses about the frequency of intake (predefined responses and open-ended responses), may be pooled to achieve the desired sample size. The common practice is to categorize open-ended responses into the predefined response categories. A problem arises when the predefined categories are noncontiguous: possible open-ended responses may fall in gaps between the predefined categories. Using simulated data modeled from frequency of intake among 1,664 controls in a lung cancer case-control study at The University of Texas M. D. Anderson Cancer Center (Houston, Texas, 2000-2005), the authors describe the effect of different categories of open-ended responses that fall in between noncontiguous, predefined response sets on estimates of the mean difference in intake and the odds ratios. A significant inflation of false positives appears when comparing mean differences of intake, while the bias in estimating odds ratios may be acceptably small. Therefore, if pooling data cannot be restricted to the same type of response, inferences should focus on odds ratio estimation to minimize bias.