JASPAR (http://jaspar.genereg.net/) is an open-access database containing manually curated, non-redundant transcription factor (TF) binding profiles for TFs across six taxonomic groups. In this 9th release, we expanded the CORE collection with 341 new profiles (148 for plants, 101 for vertebrates, 85 for urochordates, and 7 for insects), which corresponds to a 19% expansion over the previous release. We added 298 new profiles to the Unvalidated collection when no orthogonal evidence was found in the literature. All the profiles were clustered to provide familial binding profiles for each taxonomic group. Moreover, we revised the structural classification of DNA binding domains to consider plant-specific TFs. This release introduces word clouds to represent the scientific knowledge associated with each TF. We updated the genome tracks of TFBSs predicted with JASPAR profiles in eight organisms; the human and mouse TFBS predictions can be visualized as native tracks in the UCSC Genome Browser. Finally, we provide a new tool to perform JASPAR TFBS enrichment analysis in user-provided genomic regions. All the data is accessible through the JASPAR website, its associated RESTful API, the R/Bioconductor data package, and a new Python package, pyJASPAR, that facilitates serverless access to the data.

Background: Prostate cancer (PCa) has the highest incidence rates of cancers in men in western countries. Unlike several other types of cancer, PCa has few genetic drivers, which has led researchers to look for additional epigenetic and transcriptomic contributors to PCa development and progression. Especially datasets on DNA methylation, the most commonly studied epigenetic marker, have recently been measured and analysed in several PCa patient cohorts. DNA methylation is most commonly associated with downregulation of gene expression. However, positive associations of DNA methylation to gene expression have also been reported, suggesting a more diverse mechanism of epigenetic regulation. Such additional complexity could have important implications for understanding prostate cancer development but has not been studied at a genome-wide scale. Results: In this study, we have compared three sets of genome-wide single-site DNA methylation data from 870 PCa and normal tissue samples with multi-cohort gene expression data from 1117 samples, including 532 samples where DNA methylation and gene expression have been measured on the exact same samples. Genes were classified according to their corresponding methylation and expression profiles. A large group of hypermethylated genes was robustly associated with increased gene expression (UPUP group) in all three methylation datasets. These genes demonstrated distinct patterns of correlation between DNA methylation and gene expression compared to the genes showing the canonical negative association between methylation and expression (UPDOWN group). This indicates a more diversified role of DNA methylation in regulating gene expression than previously appreciated. Moreover, UPUP and UPDOWN genes were associated with different compartments-UPUP genes were related to the structures in nucleus, while UPDOWN genes were linked to extracellular features. Conclusion: We identified a robust association between hypermethylation and upregulation of gene expression when comparing samples from prostate cancer and normal tissue. These results challenge the classical view where DNA methylation is always associated with suppression of gene expression, which underlines the importance of considering corresponding expression data when assessing the downstream regulatory effect of DNA methylation.

Sequencing technologies have changed not only our approaches to classical genetics, but also the field of epigenetics. Specific methods allow scientists to identify novel genome-wide epigenetic patterns of DNA methylation down to single-nucleotide resolution. DNA methylation is the most researched epigenetic mark involved in various processes in the human cell, including gene regulation and development of diseases, such as cancer. Increasing numbers of DNA methylation sequencing datasets from human genome are produced using various platforms—from methylated DNA precipitation to the whole genome bisulfite sequencing. Many of those datasets are fully accessible for repeated analyses. Sequencing experiments have become routine in laboratories around the world, while analysis of outcoming data is still a challenge among the majority of scientists, since in many cases it requires advanced computational skills. Even though various tools are being created and published, guidelines for their selection are often not clear, especially to non-bioinformaticians with limited experience in computational analyses. Separate tools are often used for individual steps in the analysis, and these can be challenging to manage and integrate. However, in some instances, tools are combined into pipelines that are capable to complete all the essential steps to achieve the result. In the case of DNA methylation sequencing analysis, the goal of such pipeline is to map sequencing reads, calculate methylation levels, and distinguish differentially methylated positions and/or regions. The objective of this review is to describe basic principles and steps in the analysis of DNA methylation sequencing data that in particular have been used for mammalian genomes, and more importantly to present and discuss the most pronounced computational pipelines that can be used to analyze such data. We aim to provide a good starting point for scientists with limited experience in computational analyses of DNA methylation and hydroxymethylation data, and recommend a few tools that are powerful, but still easy enough to use for their own data analysis.

The molecular diversity of prostate cancer (PCa) has been demonstrated by recent genome-wide studies, proposing a significant number of different molecular markers. However, only a few of them have been transferred into clinical practice so far. The present study aimed to identify and validate novel DNA methylation biomarkers for PCa diagnosis and prognosis. Microarray-based methylome data of well-characterized cancerous and noncancerous prostate tissue (NPT) pairs was used for the initial screening. Ten protein-coding genes were selected for validation in a set of 151 PCa, 51 NPT, as well as 17 benign prostatic hyperplasia samples. The Prostate Cancer Dataset (PRAD) of The Cancer Genome Atlas (TCGA) was utilized for independent validation of our findings. Methylation frequencies of ADAMTS12, CCDC181, FILIP1L, NAALAD2, PRKCB, and ZMIZ1 were up to 91% in our study. PCa specific methylation of ADAMTS12, CCDC181, NAALAD2, and PRKCB was demonstrated by qualitative and quantitative means (all p < 0.05). In agreement with PRAD, promoter methylation of these four genes was associated with the transcript down-regulation in the Lithuanian cohort (all p < 0.05). Methylation of ADAMTS12, NAALAD2, and PRKCB was independently predictive for biochemical disease recurrence, while NAALAD2 and PRKCB increased the prognostic power of multivariate models (all p < 0.01). The present study identified methylation of ADAMTS12, NAALAD2, and PRKCB as novel diagnostic and prognostic PCa biomarkers that might guide treatment decisions in clinical practice.

Transcription factor (TF) binding to DNA is critical to transcription regulation. Although the binding properties of numerous individual TFs are well-documented, a more detailed comprehension of how TFs interact cooperatively with DNA, forming either complex or co-binding to the same region, is required. Indeed, the combinatorial binding of TFs is essential to cell differentiation, development, and response to external stimuli. We present COBIND, a novel method based on non-negative matrix factorization (NMF) to identify TF co-binding patterns automatically. COBIND applies NMF to one-hot encoded regions flanking known TF binding sites (TFBSs) to pinpoint enriched DNA patterns at fixed distances. We applied COBIND to 8,293 TFBS datasets from UniBind for 404 TFs in seven species. The method uncovered already established co-binding patterns (e.g., between POU5F1 and SOX2 or SOX17) and new co-binding configurations not yet reported in the literature and inferred through motif similarity and protein-protein interaction knowledge. Our extensive analyses across species revealed that 84% of the studied TFs share a co-binding motif with other TFs from the same structural family. The co-binding patterns captured by COBIND are likely functionally relevant as they harbor higher evolutionarily conservation than isolated TFBSs. Open chromatin data from matching human cell lines further supported the co-binding predictions. Finally, we used single-molecule footprinting data from mouse embryonic stem cells to confirm that the co-binding events captured by COBIND were likely occurring on the same DNA molecules.

No abstract

No abstract