Analysis of lifestyle and metabolic predictors of visceral obesity with Bayesian Networks

Mixed-effect Bayesian network reveals personal effects of nutrition

Nutrition experts know by their experience that people can react very differently to the same nutrition. If we could systematically quantify these differences, it would enable more personal dietary understanding and guidance. This work proposes a mixed-effect Bayesian network as a method for modeling the multivariate system of nutrition effects. Estimation of this network reveals a system of both population-wide and personal correlations between nutrients and their biological responses. Fully Bayesian estimation in the method allows managing the uncertainty in parameters and incorporating the existing nutritional knowledge into the model. The method is evaluated by modeling data from a dietary intervention study, called Sysdimet, which contains personal observations from food records and the corresponding fasting concentrations of blood cholesterol, glucose, and insulin. The model's usefulness in nutritional guidance is evaluated by predicting personally if a given diet increases or decreases future levels of concentrations. The proposed method is shown to be comparable with the well-performing Extreme Gradient Boosting (XGBoost) decision tree method in classifying the directions of concentration increases and decreases. In addition to classification, we can also predict the precise concentration level and use the biologically interpretable model parameters to understand what personal effects contribute to the concentration. We found considerable personal differences in the contributing nutrients, and while these nutritional effects are previously known at a population level, recognizing their personal differences would result in more accurate estimates and more effective nutritional guidance.

Modeling interrelationships between health behaviors in overweight breast cancer survivors: Applying Bayesian networks

Obesity and its impact on health is a multifaceted phenomenon encompassing many factors, including demographics, environment, lifestyle, and psychosocial functioning. A systems science approach, investigating these many influences, is needed to capture the complexity and multidimensionality of obesity prevention to improve health. Leveraging baseline data from a unique clinical cohort comprising 333 postmenopausal overweight or obese breast cancer survivors participating in a weight-loss trial, we applied Bayesian networks, a machine learning approach, to infer interrelationships between lifestyle factors (e.g., sleep, physical activity), body mass index (BMI), and health outcomes (biomarkers and self-reported quality of life metrics). We used bootstrap resampling to assess network stability and accuracy, and Bayesian information criteria (BIC) to compare networks. Our results identified important behavioral subnetworks. BMI was the primary pathway linking behavioral factors to glucose regulation and inflammatory markers; the BMI-biomarker link was reproduced in 100% of resampled networks. Sleep quality was a hub impacting mental quality of life and physical health with > 95% resampling reproducibility. Omission of the BMI or sleep links significantly degraded the fit of the networks. Our findings suggest potential mechanistic pathways and useful intervention targets for future trials. Using our models, we can make quantitative predictions about health impacts that would result from targeted, weight loss and/or sleep improvement interventions. Importantly, this work highlights the utility of Bayesian networks in health behaviors research.

Body adiposity indicators and cardiometabolic risk: Cross-sectional analysis in participants from the PREDIMED-Plus trial

BACKGROUND & AIMS

Excess adiposity is associated with poor cardiometabolic (CM) health. To date, several techniques and indicators have been developed to determine adiposity. We aimed to compare the ability of traditional anthropometric, as well as standard and novel DXA-derived parameters related to overall and regional adiposity, to evaluate CM risk.

METHODS

Using the cross-sectional design in the context of the PREDIMED-Plus trial, 1207 Caucasian senior men and women with overweight/obesity and metabolic syndrome (MetS) were assessed. At baseline, anthropometry- and DXA-measured parameters of central, visceral, peripheral and central-to-peripheral adiposity together with comprehensive set of CM risk factors were obtained. Partial correlations and areas under the ROC curve (AUC) were estimated to compare each adiposity measure with CM risk parameters, separately for men and women, and in the overall sample.

RESULTS

DXA-derived indicators, other than percentage of total body fat, showed stronger correlations (rho -0.172 to 0.206, p < 0.001) with CM risk than anthropometric indicators, after controlling for age, diabetes and medication use. In both sexes, DXA-derived visceral adipose tissue measures (VAT, VAT/Total fat, visceral-to-subcutaneous fat) together with lipodystrophy indicators (Trunk/Legs fat and Android/Gynoid fat) were strongly and positively correlated (p < 0.001) with glycated hemoglobin (HbA1c), the triglyceride and glucose index (TyG), triglycerides (TG), the ratio TG/HDL-cholesterol (TG/HDL-C), and were inversely related to HDL-C levels (p < 0.001). Furthermore, in AUC analyses for both sexes, VAT/Total fat showed the highest predictive ability for abnormal HbA1c levels (AUC = 0.629), VAT for TyG (AUC = 0.626), both lipodystrophy indicators for TG (AUCs = 0.556), and Trunk/Legs fat for HDL-C (AUC = 0.556) and TG/HDL-C (AUC = 0.581).

CONCLUSIONS

DXA regional adiposity measures offer advantages beyond traditional anthropometric and DXA overall adiposity indicators for CM risk assessment in senior overweight/obese subjects with MetS. In particular, in both sexes, visceral adiposity better stratifies individuals at risk for glucose abnormalities, and indicators of lipodystrophy better predict markers of dyslipidemia.

Causal inference in obesity research

Obesity is a risk factor for a plethora of severe morbidities and premature death. Most supporting evidence comes from observational studies that are prone to chance, bias and confounding. Even data on the protective effects of weight loss from randomized controlled trials will be susceptible to confounding and bias if treatment assignment cannot be masked, which is usually the case with lifestyle and surgical interventions. Thus, whilst obesity is widely considered the major modifiable risk factor for many chronic diseases, its causes and consequences are often difficult to determine. Addressing this is important, as the prevention and treatment of any disease requires that interventions focus on causal risk factors. Disease prediction, although not dependent on knowing the causes, is nevertheless enhanced by such knowledge. Here, we provide an overview of some of the barriers to causal inference in obesity research and discuss analytical approaches, such as Mendelian randomization, that can help to overcome these obstacles. In a systematic review of the literature in this field, we found: (i) probable causal relationships between adiposity and bone health/disease, cancers (colorectal, lung and kidney cancers), cardiometabolic traits (blood pressure, fasting insulin, inflammatory markers and lipids), uric acid concentrations, coronary heart disease and venous thrombosis (in the presence of pulmonary embolism), (ii) possible causal relationships between adiposity and gray matter volume, depression and common mental disorders, oesophageal cancer, macroalbuminuria, end-stage renal disease, diabetic kidney disease, nuclear cataract and gall stone disease, and (iii) no evidence for causal relationships between adiposity and Alzheimer's disease, pancreatic cancer, venous thrombosis (in the absence of pulmonary embolism), liver function and periodontitis.

Learning the local Bayesian network structure around the ZNF217 oncogene in breast tumours

In this study, we discuss and apply a novel and efficient algorithm for learning a local Bayesian network model in the vicinity of the ZNF217 oncogene from breast cancer microarray data without having to decide in advance which genes have to be included in the learning process. ZNF217 is a candidate oncogene located at 20q13, a chromosomal region frequently amplified in breast and ovarian cancer, and correlated with shorter patient survival in these cancers. To properly address the difficulties in managing complex gene interactions given our limited sample, statistical significance of edge strengths was evaluated using bootstrapping and the less reliable edges were pruned to increase the network robustness. We found that 13 out of the 35 genes associated with deregulated ZNF217 expression in breast tumours have been previously associated with survival and/or prognosis in cancers. Identifying genes involved in lipid metabolism opens new fields of investigation to decipher the molecular mechanisms driven by the ZNF217 oncogene. Moreover, nine of the 13 genes have already been identified as putative ZNF217 targets by independent biological studies. We therefore suggest that the algorithms for inferring local BNs are valuable data mining tools for unraveling complex mechanisms of biological pathways from expression data. The source code is available at http://www710.univ-lyon1.fr/∼aaussem/Software.html.