Cheng-Jian Xu

Epigenome-Wide Meta-Analysis of Methylation in Children Related to Prenatal NO2 Air Pollution Exposure

Background: Prenatal exposure to air pollution is considered to be associated with adverse effects on child health. This may partly be mediated by mechanisms related to DNA methylation. Objectives: We investigated associations between exposure to air pollution, using nitrogen dioxide (NO2) as marker, and epigenome-wide cord blood DNA methylation. Methods: We meta-analyzed the associations between NO2 exposure at residential addresses during pregnancy and cord blood DNA methylation (Illumina 450K) in four European and North American studies (n = 1,508) with subsequent look-up analyses in children ages 4 (n = 733) and 8 (n = 786) years. Additionally, we applied a literature-based candidate approach for antioxidant and anti-inflammatory genes. To assess influence of exposure at the transcriptomics level, we related mRNA expression in blood cells to NO2 exposure in 4- (n = 111) and 16-year-olds (n = 239). Results: We found epigenome-wide significant associations [false discovery rate (FDR) p < 0.05] between maternal NO2 exposure during pregnancy and DNA methylation in newborns for 3 CpG sites in mitochondria-related genes: cg12283362 (LONP1), cg24172570 (3.8 kbp upstream of HIBADH), and cg08973675 (SLC25A28). The associations with cg08973675 methylation were also significant in the older children. Further analysis of antioxidant and anti-inflammatory genes revealed differentially methylated CpGs in CAT and TPO in newborns (FDR p < 0.05). NO2 exposure at the time of biosampling in childhood had a significant impact on CAT and TPO expression. Conclusions: NO2 exposure during pregnancy was associated with differential offspring DNA methylation in mitochondria-related genes. Exposure to NO2 was also linked to differential methylation as well as expression of genes involved in antioxidant defense pathways. Citation: Gruzieva O, Xu CJ, Breton CV, Annesi-Maesano I, Antó JM, Auffray C, Ballereau S, Bellander T, Bousquet J, Bustamante M, Charles MA, de Kluizenaar Y, den Dekker HT, Duijts L, Felix JF, Gehring U, Guxens M, Jaddoe VV, Jankipersadsing SA, Merid SK, Kere J, Kumar A, Lemonnier N, Lepeule J, Nystad W, Page CM, Panasevich S, Postma D, Slama R, Sunyer J, Söderhäll C, Yao J, London SJ, Pershagen G, Koppelman GH, Melén E. 2017. Epigenome-wide meta-analysis of methylation in children related to prenatal NO2 air pollution exposure. Environ Health Perspect 125:104–110; http://dx.doi.org/10.1289/EHP36

Paving the way of systems biology and precision medicine in allergic diseases: the MeDALL success story: Mechanisms of the Development of ALLergy; EU FP7-CP-IP; Project No: 261357; 2010-2015.

MeDALL (Mechanisms of the Development of ALLergy; EU FP7‐CP‐IP; Project No: 261357; 2010–2015) has proposed an innovative approach to develop early indicators for the prediction, diagnosis, prevention and targets for therapy. MeDALL has linked epidemiological, clinical and basic research using a stepwise, large‐scale and integrative approach: MeDALL data of precisely phenotyped children followed in 14 birth cohorts spread across Europe were combined with systems biology (omics, IgE measurement using microarrays) and environmental data. Multimorbidity in the same child is more common than expected by chance alone, suggesting that these diseases share causal mechanisms irrespective of IgE sensitization. IgE sensitization should be considered differently in monosensitized and polysensitized individuals. Allergic multimorbidities and IgE polysensitization are often associated with the persistence or severity of allergic diseases. Environmental exposures are relevant for the development of allergy‐related diseases. To complement the population‐based studies in children, MeDALL included mechanistic experimental animal studies and in vitro studies in humans. The integration of multimorbidities and polysensitization has resulted in a new classification framework of allergic diseases that could help to improve the understanding of genetic and epigenetic mechanisms of allergy as well as to better manage allergic diseases. Ethics and gender were considered. MeDALL has deployed translational activities within the EU agenda.

Paving the way of systems biology and precision medicine in allergic diseases

MeDALL (Mechanisms of the Development of ALLergy; EU FP7‐CP‐IP; Project No: 261357; 2010–2015) has proposed an innovative approach to develop early indicators for the prediction, diagnosis, prevention and targets for therapy. MeDALL has linked epidemiological, clinical and basic research using a stepwise, large‐scale and integrative approach: MeDALL data of precisely phenotyped children followed in 14 birth cohorts spread across Europe were combined with systems biology (omics, IgE measurement using microarrays) and environmental data. Multimorbidity in the same child is more common than expected by chance alone, suggesting that these diseases share causal mechanisms irrespective of IgE sensitization. IgE sensitization should be considered differently in monosensitized and polysensitized individuals. Allergic multimorbidities and IgE polysensitization are often associated with the persistence or severity of allergic diseases. Environmental exposures are relevant for the development of allergy‐related diseases. To complement the population‐based studies in children, MeDALL included mechanistic experimental animal studies and in vitro studies in humans. The integration of multimorbidities and polysensitization has resulted in a new classification framework of allergic diseases that could help to improve the understanding of genetic and epigenetic mechanisms of allergy as well as to better manage allergic diseases. Ethics and gender were considered. MeDALL has deployed translational activities within the EU agenda.

Multivariate modeling of complications with data driven variable selection: Guarding against overfitting and effects of data set size

PURPOSE Multivariate modeling of complications after radiotherapy is frequently used in conjunction with data driven variable selection. This study quantifies the risk of overfitting in a data driven modeling method using bootstrapping for data with typical clinical characteristics, and estimates the minimum amount of data needed to obtain models with relatively high predictive power. MATERIALS AND METHODS To facilitate repeated modeling and cross-validation with independent datasets for the assessment of true predictive power, a method was developed to generate simulated data with statistical properties similar to real clinical data sets. Characteristics of three clinical data sets from radiotherapy treatment of head and neck cancer patients were used to simulate data with set sizes between 50 and 1000 patients. A logistic regression method using bootstrapping and forward variable selection was used for complication modeling, resulting for each simulated data set in a selected number of variables and an estimated predictive power. The true optimal number of variables and true predictive power were calculated using cross-validation with very large independent data sets. RESULTS For all simulated data set sizes the number of variables selected by the bootstrapping method was on average close to the true optimal number of variables, but showed considerable spread. Bootstrapping is more accurate in selecting the optimal number of variables than the AIC and BIC alternatives, but this did not translate into a significant difference of the true predictive power. The true predictive power asymptotically converged toward a maximum predictive power for large data sets, and the estimated predictive power converged toward the true predictive power. More than half of the potential predictive power is gained after approximately 200 samples. Our simulations demonstrated severe overfitting (a predicative power lower than that of predicting 50% probability) in a number of small data sets, in particular in data sets with a low number of events (median: 7, 95th percentile: 32). Recognizing overfitting from an inverted sign of the estimated model coefficients has a limited discriminative value. CONCLUSIONS Despite considerable spread around the optimal number of selected variables, the bootstrapping method is efficient and accurate for sufficiently large data sets, and guards against overfitting for all simulated cases with the exception of some data sets with a particularly low number of events. An appropriate minimum data set size to obtain a model with high predictive power is approximately 200 patients and more than 32 events. With fewer data samples the true predictive power decreases rapidly, and for larger data set sizes the benefit levels off toward an asymptotic maximum predictive power.

Impact of statistical learning methods on the predictive power of multivariate normal tissue complication probability models.

PURPOSE To study the impact of different statistical learning methods on the prediction performance of multivariate normal tissue complication probability (NTCP) models. METHODS AND MATERIALS In this study, three learning methods, stepwise selection, least absolute shrinkage and selection operator (LASSO), and Bayesian model averaging (BMA), were used to build NTCP models of xerostomia following radiotherapy treatment for head and neck cancer. Performance of each learning method was evaluated by a repeated cross-validation scheme in order to obtain a fair comparison among methods. RESULTS It was found that the LASSO and BMA methods produced models with significantly better predictive power than that of the stepwise selection method. Furthermore, the LASSO method yields an easily interpretable model as the stepwise method does, in contrast to the less intuitive BMA method. CONCLUSIONS The commonly used stepwise selection method, which is simple to execute, may be insufficient for NTCP modeling. The LASSO method is recommended.

Genome-wide Interaction Analysis of Air Pollution Exposure and Childhood Asthma with Functional Follow-up

Rationale: The evidence supporting an association between traffic‐related air pollution exposure and incident childhood asthma is inconsistent and may depend on genetic factors. Objectives: To identify gene‐environment interaction effects on childhood asthma using genome‐wide single‐nucleotide polymorphism (SNP) data and air pollution exposure. Identified loci were further analyzed at epigenetic and transcriptomic levels. Methods: We used land use regression models to estimate individual air pollution exposure (represented by outdoor NO2 levels) at the birth address and performed a genome‐wide interaction study for doctors’ diagnoses of asthma up to 8 years in three European birth cohorts (n = 1,534) with look‐up for interaction in two separate North American cohorts, CHS (Childrens Health Study) and CAPPS/SAGE (Canadian Asthma Primary Prevention Study/Study of Asthma, Genetics and Environment) (n = 1,602 and 186 subjects, respectively). We assessed expression quantitative trait locus effects in human lung specimens and blood, as well as associations among air pollution exposure, methylation, and transcriptomic patterns. Measurements and Main Results: In the European cohorts, 186 SNPs had an interaction P < 1 × 10−4 and a look‐up evaluation of these disclosed 8 SNPs in 4 loci, with an interaction P < 0.05 in the large CHS study, but not in CAPPS/SAGE. Three SNPs within adenylate cyclase 2 (ADCY2) showed the same direction of the interaction effect and were found to influence ADCY2 gene expression in peripheral blood (P = 4.50 × 10−4). One other SNP with P < 0.05 for interaction in CHS, rs686237, strongly influenced UDP‐Gal:betaGlcNAc &bgr;‐1,4‐galactosyltransferase, polypeptide 5 (B4GALT5) expression in lung tissue (P = 1.18 × 10−17). Air pollution exposure was associated with differential discs, large homolog 2 (DLG2) methylation and expression. Conclusions: Our results indicated that gene‐environment interactions are important for asthma development and provided supportive evidence for interaction with air pollution for ADCY2, B4GALT5, and DLG2.

Mechanisms of the Development of Allergy (MeDALL): Introducing novel concepts in allergy phenotypes

&NA; Asthma, rhinitis, and eczema are complex diseases with multiple genetic and environmental factors interlinked through IgE‐associated and non–IgE‐associated mechanisms. Mechanisms of the Development of ALLergy (MeDALL; EU FP7‐CP‐IP; project no: 261357; 2010‐2015) studied the complex links of allergic diseases at the clinical and mechanistic levels by linking epidemiologic, clinical, and mechanistic research, including in vivo and in vitro models. MeDALL integrated 14 European birth cohorts, including 44,010 participants and 160 cohort follow‐ups between pregnancy and age 20 years. Thirteen thousand children were prospectively followed after puberty by using a newly standardized MeDALL Core Questionnaire. A microarray developed for allergen molecules with increased IgE sensitivity was obtained for 3,292 children. Estimates of air pollution exposure from previous studies were available for 10,000 children. Omics data included those from historical genome‐wide association studies (23,000 children) and DNA methylation (2,173), targeted multiplex biomarker (1,427), and transcriptomic (723) studies. Using classical epidemiology and machine‐learning methods in 16,147 children aged 4 years and 11,080 children aged 8 years, MeDALL showed the multimorbidity of eczema, rhinitis, and asthma and estimated that only 38% of multimorbidity was attributable to IgE sensitization. MeDALL has proposed a new vision of multimorbidity independent of IgE sensitization, and has shown that monosensitization and polysensitization represent 2 distinct phenotypes. The translational component of MeDALL is shown by the identification of a novel allergic phenotype characterized by polysensitization and multimorbidity, which is associated with the frequency, persistence, and severity of allergic symptoms. The results of MeDALL will help integrate personalized, predictive, preventative, and participatory approaches in allergic diseases.

Association of season of birth with DNA methylation and allergic disease

Season of birth influences allergy risk; however, the biological mechanisms underlying this observation are unclear. The environment affects DNA methylation, with potentially long‐lasting effects on gene expression and disease. This study examined whether DNA methylation could underlie the association between season of birth and allergy.

Statistical Validation of Normal Tissue Complication Probability Models

PURPOSE To investigate the applicability and value of double cross-validation and permutation tests as established statistical approaches in the validation of normal tissue complication probability (NTCP) models. METHODS AND MATERIALS A penalized regression method, LASSO (least absolute shrinkage and selection operator), was used to build NTCP models for xerostomia after radiation therapy treatment of head-and-neck cancer. Model assessment was based on the likelihood function and the area under the receiver operating characteristic curve. RESULTS Repeated double cross-validation showed the uncertainty and instability of the NTCP models and indicated that the statistical significance of model performance can be obtained by permutation testing. CONCLUSION Repeated double cross-validation and permutation tests are recommended to validate NTCP models before clinical use.

Maternal BMI at the start of pregnancy and offspring epigenome-wide DNA methylation: findings from the pregnancy and childhood epigenetics (PACE) consortium

&NA; Pre‐pregnancy maternal obesity is associated with adverse offspring outcomes at birth and later in life. Individual studies have shown that epigenetic modifications such as DNA methylation could contribute. Within the Pregnancy and Childhood Epigenetics (PACE) Consortium, we meta‐analysed the association between pre‐pregnancy maternal BMI and methylation at over 450,000 sites in newborn blood DNA, across 19 cohorts (9,340 mother‐newborn pairs). We attempted to infer causality by comparing the effects of maternal versus paternal BMI and incorporating genetic variation. In four additional cohorts (1,817 mother‐child pairs), we meta‐analysed the association between maternal BMI at the start of pregnancy and blood methylation in adolescents. In newborns, maternal BMI was associated with small (<0.2% per BMI unit (1 kg/m2), P < 1.06 × 10‐7) methylation variation at 9,044 sites throughout the genome. Adjustment for estimated cell proportions greatly attenuated the number of significant CpGs to 104, including 86 sites common to the unadjusted model. At 72/86 sites, the direction of the association was the same in newborns and adolescents, suggesting persistence of signals. However, we found evidence for acausal intrauterine effect of maternal BMI on newborn methylation at just 8/86 sites. In conclusion, this well‐powered analysis identified robust associations between maternal adiposity and variations in newborn blood DNA methylation, but these small effects may be better explained by genetic or lifestyle factors than a causal intrauterine mechanism. This highlights the need for large‐scale collaborative approaches and the application of causal inference techniques in epigenetic epidemiology.