Dataset Information

Predicting gene expression using DNA methylation in three human populations.

ABSTRACT: Background:DNA methylation, an important epigenetic mark, is well known for its regulatory role in gene expression, especially the negative correlation in the promoter region. However, its correlation with gene expression across genome at human population level has not been well studied. In particular, it is unclear if genome-wide DNA methylation profile of an individual can predict her/his gene expression profile. Previous studies were mostly limited to association analyses between single CpG site methylation and gene expression. It is not known whether DNA methylation of a gene has enough prediction power to serve as a surrogate for gene expression in existing human study cohorts with DNA samples other than RNA samples. Results:We examined DNA methylation in the gene region for predicting gene expression across individuals in non-cancer tissues of three human population datasets, adipose tissue of the Multiple Tissue Human Expression Resource Projects (MuTHER), peripheral blood mononuclear cell (PBMC) from Asthma and normal control study participates, and lymphoblastoid cell lines (LCL) from healthy individuals. Three prediction models were investigated, single linear regression, multiple linear regression, and least absolute shrinkage and selection operator (LASSO) penalized regression. Our results showed that LASSO regression has superior performance among these methods. However, the prediction power is generally low and varies across datasets. Only 30 and 42 genes were found to have cross-validation R2 greater than 0.3 in the PBMC and Adipose datasets, respectively. A substantially larger number of genes (258) were identified in the LCL dataset, which was generated from a more homogeneous cell line sample source. We also demonstrated that it gives better prediction power not to exclude any CpG probe due to cross hybridization or SNP effect. Conclusion:In our three population analyses DNA methylation of CpG sites at gene region have limited prediction power for gene expression across individuals with linear regression models. The prediction power potentially varies depending on tissue, cell type, and data sources. In our analyses, the combination of LASSO regression and all probes not excluding any probe on the methylation array provides the best prediction for gene expression.

SUBMITTER: Zhong H

PROVIDER: S-EPMC6500370 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting gene expression using DNA methylation in three human populations.

Zhong Huan H Kim Soyeon S Zhi Degui D Cui Xiangqin X

PeerJ 20190501

<h4>Background</h4>DNA methylation, an important epigenetic mark, is well known for its regulatory role in gene expression, especially the negative correlation in the promoter region. However, its correlation with gene expression across genome at human population level has not been well studied. In particular, it is unclear if genome-wide DNA methylation profile of an individual can predict her/his gene expression profile. Previous studies were mostly limited to association analyses between sing ...[more]

PMID: 31106051

Similar Datasets

Project description:BackgroundLung cancer is the leading cause of cancer-related death. While cigarette smoking is the primary cause of this malignancy, risk differs across racial/ethnic groups. For the same number of cigarettes smoked, Native Hawaiians compared to whites are at greater risk and Japanese Americans are at lower risk of developing lung cancer. DNA methylation of specific CpG sites (e.g., in AHRR and F2RL3) is the most common blood epigenetic modification associated with smoking status. However, the influence of internal smoking dose, measured by urinary nicotine equivalents (NE), on DNA methylation in current smokers has not been investigated, nor has a study evaluated whether for the same smoking dose, circulating leukocyte DNA methylation patterns differ by race.MethodsWe conducted an epigenome-wide association study (EWAS) of NE in 612 smokers from three racial/ethnic groups: whites (n = 204), Native Hawaiians (n = 205), and Japanese Americans (n = 203). Genome-wide DNA methylation profiling of blood leukocyte DNA was measured using the Illumina 450K BeadChip array. Average β value, the ratio of signal from a methylated probe relative to the sum of the methylated and unmethylated probes at that CpG, was the dependent variables in linear regression models adjusting for age, sex, race (for pan-ethnic analysis), and estimated cell-type distribution.ResultsWe found that NE was significantly associated with six differentially methylated CpG sites (Bonferroni corrected p < 1.48 × 10-7): four in or near the FOXK2, PBX1, FNDC7, and FUBP3 genes and two in non-annotated genetic regions. Higher levels of NE were associated with increasing methylation beta-valuesin all six sites. For all six CpG sites, the association was only observed in Native Hawaiians, suggesting that the influence of smoking dose on DNA methylation patterns is heterogeneous across race/ethnicity (p interactions < 8.8 × 10-8). We found two additional CpG sites associated with NE in only Native Hawaiians.ConclusionsIn conclusion, internal smoking dose was associated with increased DNA methylation in circulating leukocytes at specific sites in Native Hawaiian smokers but not in white or Japanese American smokers.

Project description:BackgroundFetal sex and placental development impact pregnancy outcomes and fetal-maternal health, but the critical timepoint of placenta establishment in first trimester is understudied in human pregnancies.MethodsPregnant subjects were recruited in late first trimester (weeks 10-14) at time of chorionic villus sampling, a prenatal diagnostic test. Leftover placenta tissue was collected and stored until birth outcomes were known, then DNA and RNA were isolated from singleton, normal karyotype pregnancies resulting in live births. DNA methylation was measured with the Illumina Infinium MethylationEPIC BeadChip array (n = 56). Differential methylation analysis compared 25 females versus 31 males using a generalized linear model on 743,461 autosomal probes. Gene expression sex differences were analyzed with RNA-sequencing (n = 74). An integrated analysis was performed using linear regression to correlate gene expression and DNA methylation in 51 overlapping placentas.ResultsMethylation analysis identified 151 differentially methylated probes (DMPs) significant at false discovery rate < 0.05, including 89 (59%) hypermethylated in females. Probe cg17612569 (GABPA, ATP5J) was the most significant CpG site, hypermethylated in males. There were 11 differentially methylated regions affected by fetal sex, with transcription factors ZNF300 and ZNF311 most significantly hypermethylated in males and females, respectively. RNA-sequencing identified 152 genes significantly sexually dimorphic at false discovery rate < 0.05. The 151 DMPs were associated with 18 genes with gene downregulation (P < 0.05) in the direction of hypermethylation, including 2 genes significant at false discovery rate < 0.05 (ZNF300 and CUB and Sushi multiple domains 1, CSMD1). Both genes, as well as Family With Sequence Similarity 228 Member A (FAM228A), showed significant correlation between DNA methylation and sexually dimorphic gene expression, though FAM228A DNA methylation was less sexually dimorphic. Comparison with other sex differences studies found that cg17612569 is male-hypermethylated across gestation in placenta and in human blood up to adulthood.ConclusionsOverall, sex dimorphic differential methylation with associated differential gene expression in the first trimester placenta is small, but there remain significant genes that may be regulated through methylation leading to differences in the first trimester placenta.

Project description:Heart failure (HF) is a leading cause of morbidity and mortality in the United States and worldwide. As a multifactorial syndrome with unpredictable clinical outcomes, identifying the common molecular underpinnings that drive HF pathogenesis remains a major focus of investigation. Disruption of cardiac gene expression has been shown to mediate a common final cascade of pathological hallmarks wherein the heart reactivates numerous developmental pathways. Although the central regulatory mechanisms that drive this cardiac transcriptional reprogramming remain unknown, epigenetic contributions are likely. In the current study, we examined whether the epigenome, specifically DNA methylation, is reprogrammed in HF to potentiate a pathological shift in cardiac gene expression. To accomplish this, we used paired-end whole genome bisulfite sequencing and next-generation RNA sequencing of left ventricle tissue obtained from seven patients with end-stage HF and three nonfailing donor hearts. We found that differential methylation was localized to promoter-associated cytosine-phosphate-guanine islands, which are established regulatory regions of downstream genes. Hypermethylated promoters were associated with genes involved in oxidative metabolism, whereas promoter hypomethylation enriched glycolytic pathways. Overexpression of plasmid-derived DNA methyltransferase 3A in vitro was sufficient to lower the expression of numerous oxidative metabolic genes in H9c2 rat cardiomyoblasts, further supporting the importance of epigenetic factors in the regulation of cardiac metabolism. Last, we identified binding-site competition via hypermethylation of the nuclear respiratory factor 1 (NRF1) motif, an established upstream regulator of mitochondrial biogenesis. These preliminary observations are the first to uncover an etiology-independent shift in cardiac DNA methylation that corresponds with altered metabolic gene expression in HF.NEW & NOTEWORTHY The failing heart undergoes profound metabolic changes because of alterations in cardiac gene expression, reactivating glycolytic genes and suppressing oxidative metabolic genes. In the current study, we discover that alterations to cardiac DNA methylation encode this fetal-like metabolic gene reprogramming. We also identify novel epigenetic interference of nuclear respiratory factor 1 via hypermethylation of its downstream promoter targets, further supporting a novel contribution of DNA methylation in the metabolic remodeling of heart failure.

Dataset Information

Predicting gene expression using DNA methylation in three human populations.

Publications

Predicting gene expression using DNA methylation in three human populations.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets