DNA methylation is an important epigenetic modification involved in many biological processes and diseases. Recent developments in whole genome bisulfite sequencing (WGBS) technology have enabled genome-wide measurements of DNA methylation at single base pair resolution. Many experiments have been conducted to compare DNA methylation profiles under different biological contexts, with the goal of identifying differentially methylated regions (DMRs). Due to the high cost of WGBS experiments, many studies are still conducted without biological replicates. Methods and tools available for analyzing such data are very limited.We develop a statistical method, DSS-single, for detecting DMRs from WGBS data without replicates. We characterize the count data using a rigorous model that accounts for the spatial correlation of methylation levels, sequence depth and biological variation. We demonstrate that using information from neighboring CG sites, biological variation can be estimated accurately even without replicates. DMR detection is then carried out via a Wald test procedure. Simulations demonstrate that DSS-single has greater sensitivity and accuracy than existing methods, and an analysis of H1 versus IMR90 cell lines suggests that it also yields the most biologically meaningful results. DSS-single is implemented in the Bioconductor package DSS.

In the interphase nucleus, chromatin is organized into three-dimensional conformation to coordinate genome functions. The lamina-chromatin association is important to facilitate higher-order chromatin in mammalian cells, but its biological significances and molecular mechanisms remain poorly understood. One obstacle is that the list of lamina-associated proteins remains limited, presumably due to the inherent insolubility of lamina proteins. In this report, we identified 182 proteins associated with lamin B1 (a constitutive component of lamina) in mouse hepatocytes, by adopting virus-based proximity-dependent biotin identification. These proteins are functionally related to biological processes such as chromatin organization. As an example, we validated the association between lamin B1 and core histone macroH2A1, a histone associated with repressive chromatin. Furthermore, we mapped Lamina-associated domains (LADs) in mouse liver cells and found that boundaries of LADs are enriched for macroH2A. More interestingly, knocking-down of macroH2A1 resulted in the release of heterochromatin foci marked by histone lysine 9 trimethylation (H3K9me3) and the decondensation of global chromatin structure. However, down-regulation of lamin B1 led to redistribution of macroH2A1. Taken together, our data indicated that macroH2A1 is associated with lamina and is required to maintain chromatin architecture in mouse liver cells.

False discovery rate (FDR) control is an important tool of statistical inference in feature selection. In mass spectrometry-based metabolomics data, features can be measured at different levels of reliability and false features are often detected in untargeted metabolite profiling as chemical and/or bioinformatics noise. The traditional false discovery rate methods treat all features equally, which can cause substantial loss of statistical power to detect differentially expressed features. We propose a reliability index for mass spectrometry-based metabolomics data with repeated measurements, which is quantified using a composite measure. We then present a new method to estimate the local false discovery rate (lfdr) that incorporates feature reliability. In simulations, our proposed method achieved better balance between sensitivity and controlling false discovery, as compared to traditional lfdr estimation. We applied our method to a real metabolomics dataset and were able to detect more differentially expressed metabolites that were biologically meaningful.

Many hepatic functions including lipid metabolism, drug metabolism, and inflammatory responses are regulated in a sex-specific manner due to distinct patterns of hepatic gene expression between males and females. Regulation for the majority of these genes is under control of Nuclear Receptors (NRs). Retinoid X Receptor alpha (RXRα) is an obligate partner for multiple NRs and considered a master regulator of hepatic gene expression, yet the full extent of RXRα chromatin binding in male and female livers is unclear. ChIP-Seq analysis of RXRα and RNA Polymerase2 (Pol2) binding was performed livers of both genders and combined with microarray analysis. Mice were gavage-fed with the RXR ligand LG268 for 5 days (30 mg/kg/day) and RXRα-binding and RNA levels were determined by ChIP-qPCR and qPCR, respectively. ChIP-Seq revealed 47,845 (male) and 46,877 (female) RXRα binding sites (BS), associated with ~12,700 unique genes in livers of both genders, with 91% shared between sexes. RXRα-binding showed significant enrichment for 2227 and 1498 unique genes in male and female livers, respectively. Correlating RXRα binding strength with Pol2-binding revealed 44 genes being male-dominant and 43 female-dominant, many previously unknown to be sexually-dimorphic. Surprisingly, genes fundamental to lipid metabolism, including Scd1, Fasn, Elovl6, and Pnpla3-implicated in Fatty Liver Disease pathogenesis, were predominant in females. RXRα activation using LG268 confirmed RXRα-binding was 2–3 fold increased in female livers at multiple newly identified RXRα BS including for Pnpla3 and Elovl6, with corresponding ~10-fold and ~2-fold increases in Pnpla3 and Elovl6 RNA respectively in LG268-treated female livers, supporting a role for RXRα regulation of sexually-dimorphic responses for these genes. RXRα appears to be one of the most widely distributed transcriptional regulators in mouse liver and is engaged in determining sexually-dimorphic expression of key lipid-processing genes, suggesting novel gender- and gene-specific responses to NR-based treatments for lipid-related liver diseases.

Background: DNA methylation is a key epigenetic regulator contributing to cancer development. To understand the role of DNA methylation in tumorigenesis, it is important to investigate and compare differential methylation (DM) patterns between normal and case samples across different cancer types. However, current pan-cancer analyses call DM separately for each cancer, which suffers from lower statistical power and fails to provide a comprehensive view for patterns across cancers. Methods: In this work, we propose a rigorous statistical model, PanDM, to jointly characterize DM patterns across diverse cancer types. PanDM uses the hidden correlations in the combined dataset to improve statistical power through joint modeling. PanDM takes summary statistics from separate analyses as input and performs methylation site clustering, differential methylation detection, and pan-cancer pattern discovery. We demonstrate the favorable performance of PanDM using simulation data. We apply our model to 12 cancer methylome data collected from The Cancer Genome Atlas (TCGA) project. We further conduct ontology- and pathway-enrichment analyses to gain new biological insights into the pan-cancer DM patterns learned by PanDM. Results: PanDM outperforms two types of separate analyses in the power of DM calling in the simulation study. Application of PanDM to TCGA data reveals 37 pan-cancer DM patterns in the 12 cancer methylomes, including both common and cancer-type-specific patterns. These 37 patterns are in turn used to group cancer types. Functional ontology and biological pathways enriched in the non-common patterns not only underpin the cancer-type-specific etiology and pathogenesis but also unveil the common environmental risk factors shared by multiple cancer types. Moreover, we also identify PanDM-specific DM CpG sites that the common strategy fails to detect. Conclusions: PanDM is a powerful tool that provides a systematic way to investigate aberrant methylation patterns across multiple cancer types. Results from real data analyses suggest a novel angle for us to understand the common and specific DM patterns in different cancers. Moreover, as PanDM works on the summary statistics for each cancer type, the same framework can in principle be applied to pan-cancer analyses of other functional genomic profiles. We implement PanDM as an R package, which is freely available at http://www.sta.cuhk.edu.hk/YWei/PanDM.html.

The proposition of cancer cells in a tumor sample, named as tumor purity, is an intrinsic factor of tumor samples and has potentially great influence in variety of analyses including differential methylation, subclonal deconvolution and subtype clustering. InfiniumPurify is an integrated R package for estimating and accounting for tumor purity based on DNA methylation Infinium 450 k array data. InfiniumPurify has three main functions getPurity, InfiniumDMC and InfiniumClust, which could infer tumor purity, differential methylation analysis and tumor sample cluster accounting for estimated or user-provided tumor purities, respectively. The InfiniumPurify package provides a comprehensive analysis of tumor purity in cancer methylation research.

Antibody responses to viral infections are sustained for decades by long-lived plasma cells (LLPCs). However, LLPCs have yet to be characterized in humans. Here we used CD19, CD38, and CD138 to identify four PC subsets in human bone marrow (BM). We found that the CD19−CD38hiCD138+ subset was morphologically distinct, differentially expressed PC-associated genes and exclusively contained PCs specific for viral antigens to which the subjects had not been exposed for over 40 years. Protein sequences of measles- and mumps-specific circulating antibodies were encoded for by CD19−CD38hiCD138+ PCs in the BM. Finally, we found that CD19−CD38hiCD138+ PCs had a distinct RNA transcriptome signature and human immunoglobulin heavy chain (VH) repertoire that was relatively uncoupled from other BM PC subsets and likely represents the B cell response’s “historical record” of antigenic exposure. Thus, our studies define human LLPCs and provide a mechanism for the life-long maintenance of anti-viral antibodies in the serum.

Cerebral organoids, three-dimensional cultures that model organogenesis, provide a new platform to investigate human brain development. High cost, variability, and tissue heterogeneity limit their broad applications. Here, we developed a miniaturized spinning bioreactor (SpinΩ) to generate forebrain-specific organoids from human iPSCs. These organoids recapitulate key features of human cortical development, including progenitor zone organization, neurogenesis, gene expression, and, notably, a distinct human-specific outer radial glia cell layer. We also developed protocols for midbrain and hypothalamic organoids. Finally, we employed the forebrain organoid platform to model Zika virus (ZIKV) exposure. Quantitative analyses revealed preferential, productive infection of neural progenitors with either African or Asian ZIKV strains. ZIKV infection leads to increased cell death and reduced proliferation, resulting in decreased neuronal cell-layer volume resembling microcephaly. Together, our brain-region-specific organoids and SpinΩ provide an accessible and versatile platform for modeling human brain development and disease and for compound testing, including potential ZIKV antiviral drugs.

During the progression of pancreatic ductal adenocarcinoma (PDAC), heterogeneous subclonal populations emerge that drive primary tumor growth, regional spread, distant metastasis, and patient death. However, the genetics of metastases largely reflects that of the primary tumor in untreated patients, and PDAC driver mutations are shared by all subclones. This raises the possibility that an epigenetic process might operate during metastasis. Here we report large-scale reprogramming of chromatin modifications during the natural evolution of distant metastasis. Changes were targeted to thousands of large chromatin domains across the genome that collectively specified malignant traits, including euchromatin and large organized chromatin histone H3 lysine 9 (H3K9)-modified (LOCK) heterochromatin. Remarkably, distant metastases co-evolved a dependence on the oxidative branch of the pentose phosphate pathway (oxPPP), and oxPPP inhibition selectively reversed reprogrammed chromatin, malignant gene expression programs, and tumorigenesis. These findings suggest a model whereby linked metabolic-epigenetic programs are selected for enhanced tumorigenic fitness during the evolution of distant metastasis.

The non-human primate MPTP model of Parkinson's disease is an essential tool for translational studies. However, the currently used methodologies to produce parkinsonian monkeys do not follow unified criteria, and the applied models may often fall short of reproducing the characteristics of patients in clinical trials. Pooling of data from the parkinsonian monkeys produced in our Centers provided the opportunity to evaluate thoroughly the behavioral outcomes that may be considered for appropriate modeling in preclinical studies. We reviewed records from 108 macaques including rhesus and cynomolgus species used to model moderate to advanced parkinsonism with systemic MPTP treatment. The attained motor disability and the development of levodopa-induced dyskinesias, as primary outcomes, and the occurrence of clinical complications and instability of symptoms were all analyzed for correlations with the parameters of MPTP administration and for estimation of sample sizes. Results showed that frequently the MPTP-treated macaque can recapitulate the phenotype of patients entering clinical trials, but to produce this model consistently it is important to adapt the MPTP exposure tightly according to individual animal responses. For studies of reduced animal numbers it is also important to produce stable models, and stability of parkinsonism in macaques critically depends on reaching "marked" motor disability. The analyzed data also led to put forward recommendations for successfully producing the primate MPTP model of Parkinson's disease for translational studies.