EWAS: epigenome-wide association study software 2.0

DNA methylation variations underlie lettuce domestication and divergence

BACKGROUND

Lettuce (Lactuca sativa L.) is an economically important vegetable crop worldwide. Lettuce is believed to be domesticated from a single wild ancestor Lactuca serriola and subsequently diverged into two major morphologically distinct vegetable types: leafy lettuce and stem lettuce. However, the role of epigenetic variation in lettuce domestication and divergence remains largely unknown.

RESULTS

To understand the genetic and epigenetic basis underlying lettuce domestication and divergence, we generate single-base resolution DNA methylomes from 52 Lactuca accessions, including major lettuce cultivars and wild relatives. We find a significant increase of DNA methylation during lettuce domestication and uncover abundant epigenetic variations associated with lettuce domestication and divergence. Interestingly, DNA methylation variations specifically associated with leafy and stem lettuce are related to regulation and metabolic processes, respectively, while those associated with both types are enriched in stress responses. Moreover, we reveal that domestication-induced DNA methylation changes could influence expression levels of nearby and distal genes possibly through affecting chromatin accessibility and chromatin loop.

CONCLUSION

Our study provides population epigenomic insights into crop domestication and divergence and valuable resources for further domestication for diversity and epigenetic breeding to boost crop improvement.

Epigenome-wide methylation haplotype association analysis identified HLA-DRB1, HLA-DRB5 and HLA-DQB1 as risk factors for rheumatoid arthritis

The aim of this study was to compare nonrandom associations between physically adjacent single methylation polymorphism loci among rheumatoid arthritis (RA) and normal subjects for investigating RA-risk methylation haplotypes (meplotype). With 354 ACPA-positive RA patients and 335 normal controls selected from a case-control study based on Swedish population, we conducted the first RA epigenome-wide meplotype association study using our software EWAS2.0, mainly including (i) converted the β value to methylation genotype (menotype) data, (ii) identified methylation disequilibrium (MD) block, (iii) calculated frequent of each meplotypes in MD block and performed case-control association test and (iv) screened for RA-risk meplotypes by odd ratio (OR) and p-values. Ultimately, 545 meplotypes on 334 MD blocks were identified significantly associated with RA (p-value < .05). These meplotypes were mapped to 329 candidate genes related to RA. Subsequently, combined with gene optimization, eight RA-risk meplotypes were identified on three risk genes: HLA-DRB1, HLA-DRB5 and HLA-DQB1. Our results reported the relationship between DNA methylation pattern on HLA-DQB1 and the risk of RA for the first time, demonstrating the co-demethylation of 'cg22984282' and 'cg13423887' on HLA-DQB1 gene (meplotype UU, p-value = 2.90E - 6, OR = 1.68, 95% CI = [1.35, 2.10]) may increase the risk of RA. Our results demonstrates the potential of methylation haplotype analysis to identify RA-related genes from a new perspective and its applicability to the study of other disease.

Methylation haplotypes of the insulin gene promoter in children and adolescents with type 1 diabetes: Can a dimensionality reduction approach predict the disease?

DNA methylation of cytosine-guanine sites (CpGs) is associated with type 1 diabetes (T1D). The sequence of methylated and non-methylated sites in a specific genetic region constitutes its methyl-haplotype. The aim of the present study was to identify insulin gene promoter (IGP) methyl-haplotypes among children and adolescents with T1D and suggest a predictive model for the discrimination of cases and controls according to methyl-haplotypes. A total of 40 individuals (20 T1D) participated. The IGP region from peripheral whole blood DNA of 40 participants (20 T1D) was sequenced using next-generation sequencing, sequences were read using FASTQ files and methylation status was calculated by python-based pipeline for targeted deep bisulfite sequenced amplicons (ampliMethProfiler). Methylation profile at 10 CpG sites proximal to transcription start site of the IGP was recorded and coded as 0 for unmethylation or 1 for methylation. A single read could result in '1111111111' methyl-haplotype (all methylated), '000000000' methyl-haplotype (all unmethylated) or any other combination. Principal component analysis was applied to the generated methyl-haplotypes for dimensionality reduction, and the first three principal components were employed as features with five different classifiers (random forest, decision tree, logistic regression, Naive Bayes, support vector machine). Naive Bayes was the best-performing classifier, with 0.9 accuracy. Predictive models were evaluated using receiver operating characteristics (AUC 0.96). Methyl-haplotypes '1111111111', '1111111011', '1110111111', '1111101111' and '1110101111' were revealed to be the most significantly associated with T1D according to the dimensionality reduction method. Methylation-based biomarkers such as IGP methyl-haplotypes could serve to identify individuals at high risk for T1D.

Asymmetric variation in DNA methylation during domestication and de-domestication of rice

Hundreds of plant species have been domesticated to feed human civilization, while some crops have undergone de-domestication into agricultural weeds, threatening global food security. To understand the genetic and epigenetic basis of crop domestication and de-domestication, we generated DNA methylomes from 95 accessions of wild rice (Oryza rufipogon L.), cultivated rice (Oryza sativa L.) and weedy rice (O. sativa f. spontanea). We detected a significant decrease in DNA methylation over the course of rice domestication but observed an unexpected increase in DNA methylation through de-domestication. Notably, DNA methylation changes occurred in distinct genomic regions for these 2 opposite stages. Variation in DNA methylation altered the expression of nearby and distal genes through affecting chromatin accessibility, histone modifications, transcription factor binding, and the formation of chromatin loops, which may contribute to morphological changes during domestication and de-domestication of rice. These insights into population epigenomics underlying rice domestication and de-domestication provide resources and tools for epigenetic breeding and sustainable agriculture.

Local diversity of drought resistance and resilience in Populus tomentosa correlates with the variation of DNA methylation

Little information is known about DNA methylation variation in shaping environment-specific drought resistance and resilience for tree adaptation. In this study, we leveraged RNA sequencing and whole-genome bisulfite sequencing data to dissect the distinction of epigenetic regulation under drought stress and rewater condition of Populus tomentosa accessions from three geographical regions. We demonstrated low resistance and high resilience for accessions from South. Non-CG methylation levels in promoter regions of Southern accessions were lower than accessions from higher latitudes and negatively regulated gene expression. CHH context methylation was more sensitive to drought stress, and the geographical-specific differentially methylated regions were scarcely changed by environmental fluctuation. We identified 60 conserved hub genes within the co-expression networks that correlate with photosynthetic and stomatal morphological traits. Epigenome-wide association studies and genome-wide association studies of these 60 hub genes revealed the interdependency between genetic and epigenetic variation in GATA9 and LECRK-VIII.2, which was associated with stomatal morphology and chlorophyll content. The natural epigenetic variation in GATA9 was also faithfully transmitted to progenies in two family-based F1 populations. This study indicates a functional relationship of DNA methylation diversity with drought resistance and resilience which offers new insights into plants' local adaptation to a stressful environment.

Identification of novel meQTLs strongly associated with rheumatoid arthritis by large-scale epigenome-wide analysis

Rheumatoid arthritis (RA) is highly heritable, and previous studies have suggested that genetic variation may affect susceptibility to RA by altering epigenetic modifications (e.g. DNA methylation). Here we examined how genetic variation influences DNA methylation (DNAm) in RA by integrating individual genetic variation and DNAm data. Epigenome-wide meQTL (methylation quantitative trait loci) analysis was performed on 354 RA patients and 335 controls, scanning 30,101,744 relationships between 62 SNPs and 485,512 DNA methylation sites. Two regulatory relationship pairs (FDR < 0.05) showed very strong associations with RA risk. One was rs10796216-cg00475509, and the DNAm decreased by 0.0168 per addition of allele rs10796216-A. The other was rs6546473-cg13358873, for which a 0.0365 reduction of DNAm at cg13358873 was observed for each addition of allele rs6546473-A, and lower DNAm was found to be significantly associated with RA risk (P = 2.0407e-28). Moreover, both pairs of meQTL showed a strong regulatory relationship only in RA samples, so they can be subsequently considered as risk markers for RA. In conclusion, our integrated analysis of genetic and epigenetic variation suggests that genetic variation may affect the risk of RA by regulating DNA methylation levels. Alterations of DNAm at cg00475509 and cg13358873 loci conferred by rs10796216-A and rs6546473-A allele may suggest a potential risk for RA. Our results deepen our understanding of the genetic and epigenetic mechanisms of RA and provide novel associations that can be further investigated in future studies.

RABC: Rheumatoid Arthritis Bioinformatics Center

Advances in sequencing technologies have led to the rapid growth of multi-omics data on rheumatoid arthritis (RA). However, a comprehensive database that systematically collects and classifies the scattered data is still lacking. Here, we developed the Rheumatoid Arthritis Bioinformatics Center (RABC, http://www.onethird-lab.com/RABC/), the first multi-omics data resource platform (data hub) for RA. There are four categories of data in RABC: (i) 175 multi-omics sample sets covering transcriptome, epigenome, genome, and proteome; (ii) 175 209 differentially expressed genes (DEGs), 105 differentially expressed microRNAs (DEMs), 18 464 differentially DNA methylated (DNAm) genes, 1 764 KEGG pathways, 30 488 GO terms, 74 334 SNPs, 242 779 eQTLs, 105 m6A-SNPs and 18 491 669 meta-mQTLs; (iii) prior knowledge on seven types of RA molecular markers from nine public and credible databases; (iv) 127 073 literature information from PubMed (from 1972 to March 2022). RABC provides a user-friendly interface for browsing, searching and downloading these data. In addition, a visualization module also supports users to generate graphs of analysis results by inputting personalized parameters. We believe that RABC will become a valuable resource and make a significant contribution to the study of RA.

Multi-Omics Profiling Approach to Asthma: An Evolving Paradigm

Asthma is a complex multifactorial and heterogeneous respiratory disease. Although genetics is a strong risk factor of asthma, external and internal exposures and their interactions with genetic factors also play important roles in the pathophysiology of asthma. Over the past decades, the application of high-throughput omics approaches has emerged and been applied to the field of asthma research for screening biomarkers such as genes, transcript, proteins, and metabolites in an unbiased fashion. Leveraging large-scale studies representative of diverse population-based omics data and integrating with clinical data has led to better profiling of asthma risk. Yet, to date, no omic-driven endotypes have been translated into clinical practice and management of asthma. In this article, we provide an overview of the current status of omics studies of asthma, namely, genomics, transcriptomics, epigenomics, proteomics, exposomics, and metabolomics. The current development of the multi-omics integrations of asthma is also briefly discussed. Biomarker discovery following multi-omics profiling could be challenging but useful for better disease phenotyping and endotyping that can translate into advances in asthma management and clinical care, ultimately leading to successful precision medicine approaches.

EWAS Open Platform: integrated data, knowledge and toolkit for epigenome-wide association study

Epigenome-Wide Association Study (EWAS) has become a standard strategy to discover DNA methylation variation of different phenotypes. Since 2018, we have developed EWAS Atlas and EWAS Data Hub to integrate a growing volume of EWAS knowledge and data, respectively. Here, we present EWAS Open Platform (https://ngdc.cncb.ac.cn/ewas) that includes EWAS Atlas, EWAS Data Hub and the newly developed EWAS Toolkit. In the current implementation, EWAS Open Platform integrates 617 018 high-quality EWAS associations from 910 publications, covering 51 phenotypes, 275 diseases and 104 environmental factors. It also provides well-normalized DNA methylation array data and the corresponding metadata from 115 852 samples, which involve 707 tissues, 218 cell lines and 528 diseases. Taking advantage of integrated knowledge and data in EWAS Atlas and EWAS Data Hub, EWAS Open Platform equips with EWAS Toolkit, a powerful one-stop site for EWAS enrichment, annotation, and knowledge network construction and visualization. Collectively, EWAS Open Platform provides open access to EWAS knowledge, data and toolkit and thus bears great utility for a broader range of relevant research.

Brain Immunoinformatics: A Symmetrical Link between Informatics, Wet Lab and the Clinic

Breakthrough advances in informatics over the last decade have thoroughly influenced the field of immunology. The intermingling of machine learning with wet lab applications and clinical results has hatched the newly defined immunoinformatics society. Immunoinformatics of the central neural system, referred to as neuroimmunoinformatics (NII), investigates symmetrical and asymmetrical interactions of the brain-immune interface. This interdisciplinary overview on NII is addressed to bioscientists and computer scientists. We delineate the dominating trajectories and field-shaping achievements and elaborate on future directions using bridging language and terminology. Computation, varying from linear modeling to complex deep learning approaches, fuels neuroimmunology through three core directions. Firstly, by providing big-data analysis software for high-throughput methods such as next-generation sequencing and genome-wide association studies. Secondly, by designing models for the prediction of protein morphology, functions, and symmetrical and asymmetrical protein–protein interactions. Finally, NII boosts the output of quantitative pathology by enabling the automatization of tedious processes such as cell counting, tracing, and arbor analysis. The new classification of microglia, the brain’s innate immune cells, was an NII achievement. Deep sequencing classifies microglia in “sensotypes” to accurately describe the versatility of immune responses to physiological and pathological challenges, as well as to experimental conditions such as xenografting and organoids. NII approaches complex tasks in the brain-immune interface, recognizes patterns and allows for hypothesis-free predictions with ultimate targeted individualized treatment strategies, and personalizes disease prognosis and treatment response.

Ten Years of EWAS

Epigenome-wide association study (EWAS) has been applied to analyze DNA methylation variation in complex diseases for a decade, and epigenome as a research target has gradually become a hot topic of current studies. The DNA methylation microarrays, next-generation, and third-generation sequencing technologies have prepared a high-quality platform for EWAS. Here, the progress of EWAS research is reviewed, its contributions to clinical applications, and mainly describe the achievements of four typical diseases. Finally, the challenges encountered by EWAS and make bold predictions for its future development are presented.

Mergeomics 2.0: a web server for multi-omics data integration to elucidate disease networks and predict therapeutics

The Mergeomics web server is a flexible online tool for multi-omics data integration to derive biological pathways, networks, and key drivers important to disease pathogenesis and is based on the open source Mergeomics R package. The web server takes summary statistics of multi-omics disease association studies (GWAS, EWAS, TWAS, PWAS, etc.) as input and features four functions: Marker Dependency Filtering (MDF) to correct for known dependency between omics markers, Marker Set Enrichment Analysis (MSEA) to detect disease relevant biological processes, Meta-MSEA to examine the consistency of biological processes informed by various omics datasets, and Key Driver Analysis (KDA) to identify essential regulators of disease-associated pathways and networks. The web server has been extensively updated and streamlined in version 2.0 including an overhauled user interface, improved tutorials and results interpretation for each analytical step, inclusion of numerous disease GWAS, functional genomics datasets, and molecular networks to allow for comprehensive omics integrations, increased functionality to decrease user workload, and increased flexibility to cater to user-specific needs. Finally, we have incorporated our newly developed drug repositioning pipeline PharmOmics for prediction of potential drugs targeting disease processes that were identified by Mergeomics. Mergeomics is freely accessible at http://mergeomics.research.idre.ucla.edu and does not require login.

The EpiDiverse Plant Epigenome-Wide Association Studies (EWAS) Pipeline

Bisulfite sequencing is a widely used technique for determining DNA methylation and its relationship with epigenetics, genetics, and environmental parameters. Various techniques were implemented for epigenome-wide association studies (EWAS) to reveal meaningful associations; however, there are only very few plant studies available to date. Here, we developed the EpiDiverse EWAS pipeline and tested it using two plant datasets, from P. abies (Norway spruce) and Q. lobata (valley oak). Hence, we present an EWAS implementation tested for non-model plant species and describe its use.

Mining the Selective Remodeling of DNA Methylation in Promoter Regions to Identify Robust Gene-Level Associations With Phenotype

Epigenetics is an essential biological frontier linking genetics to the environment, where DNA methylation is one of the most studied epigenetic events. In recent years, through the epigenome-wide association study (EWAS), researchers have identified thousands of phenotype-related methylation sites. However, the overlaps of identified phenotype-related DNA methylation sites between various studies are often quite small, and it might be due to the fact that methylation remodeling has a certain degree of randomness within the genome. Thus, the identification of robust gene-phenotype associations is crucial to interpreting pathogenesis. How to integrate the methylation values of different sites on the same gene and to mine the DNA methylation at the gene level remains a challenge. A recent study found that the DNA methylation difference of the gene body and promoter region has a strong correlation with gene expression. In this study, we proposed a Statistical difference of DNA Methylation between Promoter and Other Body Region (SIMPO) algorithm to extract DNA methylation values at the gene level. First, by choosing to smoke as an environmental exposure factor, our method led to significant improvements in gene overlaps (from 5 to 17%) between different datasets. In addition, the biological significance of phenotype-related genes identified by SIMPO algorithm is comparable to that of the traditional probe-based methods. Then, we selected two disease contents (e.g., insulin resistance and Parkinson's disease) to show that the biological efficiency of disease-related gene identification increased from 15.43 to 44.44% (p-value = 1.20e-28). In summary, our results declare that mining the selective remodeling of DNA methylation in promoter regions can identify robust gene-level associations with phenotype, and the characteristic remodeling of a given gene's promoter region can reflect the essence of disease.

Predicting lung adenocarcinoma disease progression using methylation-correlated blocks and ensemble machine learning classifiers

Applying the knowledge that methyltransferases and demethylases can modify adjacent cytosine-phosphorothioate-guanine (CpG) sites in the same DNA strand, we found that combining multiple CpGs into a single block may improve cancer diagnosis. However, survival prediction remains a challenge. In this study, we developed a pipeline named "stacked ensemble of machine learning models for methylation-correlated blocks" (EnMCB) that combined Cox regression, support vector regression (SVR), and elastic-net models to construct signatures based on DNA methylation-correlated blocks for lung adenocarcinoma (LUAD) survival prediction. We used methylation profiles from the Cancer Genome Atlas (TCGA) as the training set, and profiles from the Gene Expression Omnibus (GEO) as validation and testing sets. First, we partitioned the genome into blocks of tightly co-methylated CpG sites, which we termed methylation-correlated blocks (MCBs). After partitioning and feature selection, we observed different diagnostic capacities for predicting patient survival across the models. We combined the multiple models into a single stacking ensemble model. The stacking ensemble model based on the top-ranked block had the area under the receiver operating characteristic curve of 0.622 in the TCGA training set, 0.773 in the validation set, and 0.698 in the testing set. When stratified by clinicopathological risk factors, the risk score predicted by the top-ranked MCB was an independent prognostic factor. Our results showed that our pipeline was a reliable tool that may facilitate MCB selection and survival prediction.

Imputing missing RNA-sequencing data from DNA methylation by using a transfer learning-based neural network

BACKGROUND

Gene expression plays a key intermediate role in linking molecular features at the DNA level and phenotype. However, owing to various limitations in experiments, the RNA-seq data are missing in many samples while there exist high-quality of DNA methylation data. Because DNA methylation is an important epigenetic modification to regulate gene expression, it can be used to predict RNA-seq data. For this purpose, many methods have been developed. A common limitation of these methods is that they mainly focus on a single cancer dataset and do not fully utilize information from large pan-cancer datasets.

RESULTS

Here, we have developed a novel method to impute missing gene expression data from DNA methylation data through a transfer learning-based neural network, namely, TDimpute. In the method, the pan-cancer dataset from The Cancer Genome Atlas (TCGA) was utilized for training a general model, which was then fine-tuned on the specific cancer dataset. By testing on 16 cancer datasets, we found that our method significantly outperforms other state-of-the-art methods in imputation accuracy with a 7-11% improvement under different missing rates. The imputed gene expression was further proved to be useful for downstream analyses, including the identification of both methylation-driving and prognosis-related genes, clustering analysis, and survival analysis on the TCGA dataset. More importantly, our method was indicated to be useful for general purposes by an independent test on the Wilms tumor dataset from the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) project.

CONCLUSIONS

TDimpute is an effective method for RNA-seq imputation with limited training samples.

EWASdb: epigenome-wide association study database

DNA methylation, the most intensively studied epigenetic modification, plays an important role in understanding the molecular basis of diseases. Furthermore, epigenome-wide association study (EWAS) provides a systematic approach to identify epigenetic variants underlying common diseases/phenotypes. However, there is no comprehensive database to archive the results of EWASs. To fill this gap, we developed the EWASdb, which is a part of 'The EWAS Project', to store the epigenetic association results of DNA methylation from EWASs. In its current version (v 1.0, up to July 2018), the EWASdb has curated 1319 EWASs associated with 302 diseases/phenotypes. There are three types of EWAS results curated in this database: (i) EWAS for single marker; (ii) EWAS for KEGG pathway and (iii) EWAS for GO (Gene Ontology) category. As the first comprehensive EWAS database, EWASdb has been searched or downloaded by researchers from 43 countries to date. We believe that EWASdb will become a valuable resource and significantly contribute to the epigenetic research of diseases/phenotypes and have potential clinical applications. EWASdb is freely available at http://www.ewas.org.cn/ewasdb or http://www.bioapp.org/ewasdb.

Genome-wide association studies of severe P. falciparum malaria susceptibility: progress, pitfalls and prospects

BACKGROUND

P. falciparum malaria has been recognized as one of the prominent evolutionary selective forces of human genome that led to the emergence of multiple host protective alleles. A comprehensive understanding of the genetic bases of severe malaria susceptibility and resistance can potentially pave ways to the development of new therapeutics and vaccines. Genome-wide association studies (GWASs) have recently been implemented in malaria endemic areas and identified a number of novel association genetic variants. However, there are several open questions around heritability, epistatic interactions, genetic correlations and associated molecular pathways among others. Here, we assess the progress and pitfalls of severe malaria susceptibility GWASs and discuss the biology of the novel variants.

RESULTS

We obtained all severe malaria susceptibility GWASs published thus far and accessed GWAS dataset of Gambian populations from European Phenome Genome Archive (EGA) through the MalariaGen consortium standard data access protocols. We noticed that, while some of the well-known variants including HbS and ABO blood group were replicated across endemic populations, only few novel variants were convincingly identified and their biological functions remain to be understood. We estimated SNP-heritability of severe malaria at 20.1% in Gambian populations and showed how advanced statistical genetic analytic methods can potentially be implemented in malaria susceptibility studies to provide useful functional insights.

CONCLUSIONS

The ultimate goal of malaria susceptibility study is to discover a novel causal biological pathway that provide protections against severe malaria; a fundamental step towards translational medicine such as development of vaccine and new therapeutics. Beyond singe locus analysis, the future direction of malaria susceptibility requires a paradigm shift from single -omics to multi-stage and multi-dimensional integrative functional studies that combines multiple data types from the human host, the parasite, the mosquitoes and the environment. The current biotechnological and statistical advances may eventually lead to the feasibility of systems biology studies and revolutionize malaria research.

Landscape of SNPs-mediated lncRNA structural variations and their implication in human complex diseases

An increasing number of functional studies shows that long noncoding RNAs (lncRNAs) are involved in many aspects of cellular physiology and fulfills a wide variety of regulatory roles at almost every stage of gene expression. A major feature of lncRNAs is the highly folded modular domains in transcripts. With improved modeling and definition, it is now feasible to explore and gain novel insights into the structural-functional relationship of lncRNAs and their association with complex human diseases. In this study, we utilized an automatic computational pipeline to scan lncRNA architecture at the genome-wide scale and to obtain a landscape of functional domains. An accurate alignment algorithm was performed to identify 40 triple pairs between single-nucleotide polymorphisms (SNPs), lncRNAs and diseases. In order to detect the potential contribution of a lncRNA's modular character, we estimated and evaluated structural rearrangements, which were derived from disease-associated SNPs. In addition, we focused on annotating and comparing the global and local heterogeneity of the wild-type and mutant lncRNAs. Assessing lncRNA architecture has yielded how variations in structured regions impact the molecular mechanisms of lncRNAs and how SNPs disturb binding and recruiting ability. These observations are the first glimpse of the 'lncRNA structurome' and make it possible to robustly explore and assemble intricate space conformation and their stress variation. This result also successfully demonstrates that lncRNA transcripts contain a complex structural landscape and highlights the proposed contribution of lncRNA structure in controlling RNA functions and disease mechanisms.