A powerful and flexible weighted distance-based method incorporating interactions between DNA methylation and environmental factors on health outcomes

MOTIVATION Deoxyribonucleic acid (DNA) methylation plays a crucial role in human health. Studies have demonstrated associations between DNA methylation and environmental factors with evidence also supporting the idea that DNA methylation may modify the risk of environmental factors on health outcomes. However, due to high dimensionality and low study power, current studies usually focus on finding differential methylation on health outcomes at CpG level or gene level combining multiple CpGs and/or finding environmental effects on health outcomes but ignoring their interactions on health outcomes. Here we introduce the idea of a pseudo-data matrix constructed with cross-product terms between CpGs and environmental factors that are able to capture their interactions. We then develop a powerful and flexible weighted distance-based method with the pseudo-data matrix where association strength was used as weights on CpGs, environmental factors and their interactions to up-weight signals and down-weight noises in distance calculations. RESULTS We compared the power of this novel approach and several comparison methods in simulated datasets and the Mothers and Newborns birth cohort of the Columbia Center for Children's Environmental Health to determine whether prenatal polycyclic aromatic hydrocarbons interacts with DNA methylation in association with Attention Deficit Hyperactivity Disorder and Mental Development Index at age 3. AVAILABILITY AND IMPLEMENTATION An R code for the proposed method Dw-M-E-int together with a tutorial and a sample dataset is available for downloading from http://www.columbia.edu/∼sw2206/softwares.htm. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.

[1]  M. Ehrlich,et al.  DNA methylation in cancer: too much, but also too little , 2002, Oncogene.

[2]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[3]  G. Arbanas.  Diagnostic and Statistical Manual of Mental Disorders (DSM-5) , 2015 .

[4]  Lode Godderis,et al.  Placental DNA hypomethylation in association with particulate air pollution in early life , 2013, Particle and Fibre Toxicology.

[5]  Patrick De Boever,et al.  Lower Placental Leptin Promoter Methylation in Association with Fine Particulate Matter Air Pollution during Pregnancy and Placental Nitrosative Stress at Birth in the ENVIRONAGE Cohort , 2016, Environmental health perspectives.

[6]  A. Feinberg.  Phenotypic plasticity and the epigenetics of human disease , 2007, Nature.

[7]  Pablo Cingolani,et al.  Lead exposure induces changes in 5-hydroxymethylcytosine clusters in CpG islands in human embryonic stem cells and umbilical cord blood , 2015, Epigenetics.

[8]  J. Mill,et al.  Molecular studies of major depressive disorder: the epigenetic perspective , 2007, Molecular Psychiatry.

[9]  Sun-Chong Wang,et al.  Epigenomic profiling reveals DNA-methylation changes associated with major psychosis. , 2008, American journal of human genetics.

[10]  Arturas Petronis,et al.  Pre- and peri-natal environmental risks for attention-deficit hyperactivity disorder (ADHD): the potential role of epigenetic processes in mediating susceptibility. , 2008, Journal of child psychology and psychiatry, and allied disciplines.

[11]  Matthew A. Zapala,et al.  Multivariate regression analysis of distance matrices for testing associations between gene expression patterns and related variables , 2006, Proceedings of the National Academy of Sciences.

[12]  Susan K. Murphy,et al.  Maternal blood lead concentrations, DNA methylation of MEG3 DMR regulating the DLK1/MEG3 imprinted domain and early growth in a multiethnic cohort , 2016, Environmental epigenetics.

[13]  Chunyang Liao,et al.  In utero bisphenol A concentration, metabolism, and global DNA methylation across matched placenta, kidney, and liver in the human fetus. , 2015, Chemosphere.

[14]  Takao K Hensch,et al.  Prenatal antidepressant exposure associated with CYP2E1 DNA methylation change in neonates , 2015, Epigenetics.

[15]  P. M. Das,et al.  DNA methylation and cancer. , 2004, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[16]  Frederica P. Perera,et al.  Prenatal Exposure to Polycyclic Aromatic Hydrocarbons, Benzo[a]pyrene–DNA Adducts, and Genomic DNA Methylation in Cord Blood , 2012, Environmental health perspectives.

[17]  M. Esteller,et al.  DNA methylation and cancer. , 2010, Advances in genetics.

[18]  Richard C. McEachin,et al.  Bisphenol A-associated alterations in genome-wide DNA methylation and gene expression patterns reveal sequence-dependent and non-monotonic effects in human fetal liver , 2015, Environmental epigenetics.

[19]  Mary Beth Terry,et al.  Polycyclic aromatic hydrocarbon (PAH)–DNA adducts and breast cancer: modification by gene promoter methylation in a population-based study , 2015, Cancer Causes & Control.

[20]  Deliang Tang,et al.  Relation of DNA Methylation of 5′-CpG Island of ACSL3 to Transplacental Exposure to Airborne Polycyclic Aromatic Hydrocarbons and Childhood Asthma , 2009, PloS one.

[21]  Robin M Whyatt,et al.  Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population. , 2003, Environmental health perspectives.

[22]  Manolis Kellis,et al.  Alzheimery's disease pathology is associated with early alterations in brain DNA methylation at ANK1, BIN1, RHBDF2 and other loci , 2014, Nature Neuroscience.

[23]  Eric J. Nestler,et al.  Epigenetic mechanisms of drug addiction , 2014, Neuropharmacology.

[24]  Qing Nie,et al.  Cell lineage and communication network inference via optimization for single-cell transcriptomics , 2019, Nucleic acids research.

[25]  J. Gent,et al.  An environmental epigenetic study of ADRB2 5′‐UTR methylation and childhood asthma severity , 2012, Clinical and experimental allergy : journal of the British Society for Allergy and Clinical Immunology.

[26]  Xiao Zhang,et al.  Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis , 2010, BMC Bioinformatics.

[27]  N. C. Schanen,et al.  Epigenetics of autism spectrum disorders. , 2006, Human molecular genetics.

[28]  M. Rutter,et al.  Severe psychosocial deprivation in early childhood is associated with increased DNA methylation across a region spanning the transcription start site of CYP2E1 , 2016, Translational psychiatry.

[29]  J. Herman,et al.  Cancer as an epigenetic disease: DNA methylation and chromatin alterations in human tumours , 2002, The Journal of pathology.

[30]  Rolf U Halden,et al.  Prenatal mercury concentration is associated with changes in DNA methylation at TCEANC2 in newborns. , 2015, International journal of epidemiology.

[31]  Andrew E Teschendorff,et al.  Detection of epigenetic field defects using a weighted epigenetic distance-based method , 2018, Nucleic acids research.

[32]  Marie-France Hivert,et al.  Persistent DNA methylation changes associated with prenatal mercury exposure and cognitive performance during childhood , 2017, Scientific Reports.

[33]  Frederica P. Perera,et al.  Effect of Prenatal Exposure to Airborne Polycyclic Aromatic Hydrocarbons on Neurodevelopment in the First 3 Years of Life among Inner-City Children , 2006, Environmental health perspectives.