Genomic Prediction Accounting for Residual Heteroskedasticity

Modeling of an integrative prototype based on genetic, phenotypic, and environmental information for personalized prescription of energy-restricted diets in overweight/obese subjects

BACKGROUND

Interindividual variability in weight loss and metabolic responses depends upon interactions between genetic, phenotypic, and environmental factors.

OBJECTIVE

We aimed to model an integrative (nutri) prototype based on genetic, phenotypic, and environmental information for the personalized prescription of energy-restricted diets with different macronutrient distribution.

METHODS

A 4-mo nutritional intervention was conducted in 305 overweight/obese volunteers involving 2 energy-restricted diets (30% restriction) with different macronutrient distribution: a moderately high-protein (MHP) diet (30% proteins, 30% lipids, and 40% carbohydrates) and a low-fat (LF) diet (22% lipids, 18% proteins, and 60% carbohydrates). A total of 201 subjects with good dietary adherence were genotyped for 95 single nucleotide polymorphisms (SNPs) related to energy homeostasis. Genotyping was performed by targeted next-generation sequencing. Two weighted genetic risk scores for the MHP (wGRS1) and LF (wGRS2) diets were computed using statistically relevant SNPs. Multiple linear regression models were performed to estimate percentage BMI decrease depending on the dietary macronutrient composition.

RESULTS

After energy restriction, both the MHP and LF diets induced similar significant decreases in adiposity, body composition, and blood pressure, and improved the lipid profile. Furthermore, statistically relevant differences in anthropometric and biochemical markers depending on sex and age were found. BMI decrease in the MHP diet was best predicted at ∼28% (optimism-corrected adjusted R2 = 0.279) by wGRS1 and age, whereas wGRS2 and baseline energy intake explained ∼29% (optimism-corrected adjusted R2 = 0.287) of BMI decrease variability in the LF diet. The incorporation of these predictive models into a decision algorithm allowed the personalized prescription of the MHP and LF diets.

CONCLUSIONS

Different genetic, phenotypic, and exogenous factors predict BMI decreases depending on the administration of a hypocaloric MHP diet or an LF diet. This holistic approach may help to personalize dietary advice for the management of excessive body weight using precision nutrition variables.This trial was registered at clinicaltrials.gov as NCT02737267.

Efficient Estimation of Marker Effects in Plant Breeding

The evaluation of prediction machines is an important step for a successful implementation of genomic-enabled selection in plant breeding. Computation time and predictive ability constitute key metrics to determine the methodology utilized for the consolidation of genomic prediction pipeline. This study introduces two methods designed to couple high prediction accuracy with efficient computational performance: 1) a non-MCMC method to estimate marker effects with a Laplace prior; and 2) an iterative framework that allows solving whole-genome regression within mixed models with replicated observations in a single-stage. The investigation provides insights on predictive ability and marker effect estimates. Various genomic prediction techniques are compared based on cross-validation, assessing predictions across and within family. Properties of quantitative trait loci detection and single-stage method were evaluated on simulated plot-level data from unbalanced data structures. Estimation of marker effects by the new model is compared to a genome-wide association analysis and whole-genome regression methods. The single-stage approach is compared to a GBLUP fitted via restricted maximum likelihood, and a two-stages approaches where genetic values fit a whole-genome regression. The proposed framework provided high computational efficiency, robust prediction across datasets, and accurate estimation of marker effects.

Models Integrating Genetic and Lifestyle Interactions on Two Adiposity Phenotypes for Personalized Prescription of Energy-Restricted Diets With Different Macronutrient Distribution

Aim: To analyze the influence of genetics and interactions with environmental factors on adiposity outcomes [waist circumference reduction (WCR) and total body fat loss (TFATL)] in response to energy-restricted diets in subjects with excessive body weight.

Materials and Methods: Two hypocaloric diets (30% energy restriction) were prescribed to overweight/obese subjects during 16 weeks, which had different targeted macronutrient distribution: a low-fat (LF) diet (22% energy from lipids) and a moderately high-protein (MHP) diet (30% energy from proteins). At the end of the trial, a total of 201 participants (LF diet = 105; MHP diet = 96) who presented good/regular dietary adherence were genotyped for 95 single nucleotide polymorphisms (SNPs) previously associated with weight loss through next-generation sequencing from oral samples. Four unweighted (uGRS) and four weighted (wGRS) genetic risk scores were computed using statistically relevant SNPs for each outcome by diet. Predictions of WCR and TFATL by diet were modeled through recognized multiple linear regression models including genetic (single SNPs, uGRS, and wGRS), phenotypic (age, sex, and WC, or TFAT at baseline), and environment variables (physical activity level and energy intake at baselines) as well as eventual interactions between genes and environmental factors.

Results: Overall, 26 different SNPs were associated with differential adiposity outcomes, 9 with WCR and 17 with TFATL, most of which were specific for each dietary intervention. In addition to conventional predictors (age, sex, lifestyle, and adiposity status at baseline), the calculated uGRS/wGRS and interactions with environmental factors were major contributors of adiposity responses. Thus, variances in TFATL-LF diet, TFATL-MHP diet, WCR-LF diet, and WCR-MHP diet were predicted by approximately 38% (optimism-corrected adj. R² = 0.3792), 32% (optimism-corrected adj. R² = 0.3208), 22% (optimism-corrected adj. R² = 0.2208), and 21% (optimism-corrected adj. R² = 0.2081), respectively.

Conclusions: Different genetic variants and interactions with environmental factors modulate the differential individual responses to MHP and LF dietary interventions. These insights and models may help to optimize personalized nutritional strategies for modeling the prevention and management of excessive adiposity through precision nutrition approaches taking into account not only genetic information but also the lifestyle/clinical factors that interplay in addition to age and sex.

Extensions of BLUP Models for Genomic Prediction in Heterogeneous Populations: Application in a Diverse Switchgrass Sample

Genomic prediction is a useful tool to accelerate genetic gain in selection using DNA marker information. However, this technology typically relies on standard prediction procedures, such as genomic BLUP, that are not designed to accommodate population heterogeneity resulting from differences in marker effects across populations. In this study, we assayed different prediction procedures to capture marker-by-population interactions in genomic prediction models. Prediction procedures included genomic BLUP and two kernel-based extensions of genomic BLUP which explicitly accounted for population heterogeneity. To model population heterogeneity, dissemblance between populations was either depicted by a unique coefficient (as previously reported), or a more flexible function of genetic distance between populations (proposed herein). Models under investigation were applied in a diverse switchgrass sample under two validation schemes: whole-sample calibration, where all individuals except selection candidates are included in the calibration set, and cross-population calibration, where the target population is entirely excluded from the calibration set. First, we showed that using fixed effects, from principal components or putative population groups, appeared detrimental to prediction accuracy, especially in cross-population calibration. Then we showed that modeling population heterogeneity by our proposed procedure resulted in highly significant improvements in model fit. In such cases, gains in accuracy were often positive. These results suggest that population heterogeneity may be parsimoniously captured by kernel methods. However, in cases where improvement in model fit by our proposed procedure is null-to-moderate, ignoring heterogeneity should probably be preferred due to the robustness and simplicity of the standard genomic BLUP model.