Dataset Information

MethRaFo: MeDIP-seq methylation estimate using a Random Forest Regressor.

ABSTRACT: Motivation:Profiling of genome wide DNA methylation is now routinely performed when studying development, cancer and several other biological processes. Although Whole genome Bisulfite Sequencing provides high-quality methylation measurements at the resolution of nucleotides, it is relatively costly and so several studies have used alternative methods for such profiling. One of the most widely used low cost alternatives is MeDIP-Seq. However, MeDIP-Seq is biased for CpG enriched regions and thus its results need to be corrected in order to determine accurate methylation levels. Results:Here we present a method for correcting MeDIP-Seq results based on Random Forest regression. Applying the method to real data from several different tissues (brain, cortex, penis) we show that it achieves almost 4 fold decrease in run time while increasing accuracy by as much as 20% over prior methods developed for this task. Availability and implementation:MethRaFo is freely available as a python package (with a R wrapper) at https://github.com/phoenixding/methrafo. Contact:zivbj@cs.cmu.edu. Supplementary information:Supplementary data are available at Bioinformatics online.

SUBMITTER: Ding J

PROVIDER: S-EPMC5860172 | biostudies-literature | 2017 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

MethRaFo: MeDIP-seq methylation estimate using a Random Forest Regressor.

Ding Jun J Bar-Joseph Ziv Z

Bioinformatics (Oxford, England) 20171101 21

<h4>Motivation</h4>Profiling of genome wide DNA methylation is now routinely performed when studying development, cancer and several other biological processes. Although Whole genome Bisulfite Sequencing provides high-quality methylation measurements at the resolution of nucleotides, it is relatively costly and so several studies have used alternative methods for such profiling. One of the most widely used low cost alternatives is MeDIP-Seq. However, MeDIP-Seq is biased for CpG enriched regions ...[more]

PMID: 29036558

Similar Datasets

Project description:BackgroundDNA methylation is an important regulator of gene expression and chromatin structure. Methylated DNA immunoprecipitation sequencing (MeDIP-Seq) is commonly used to identify regions of DNA methylation in eukaryotic genomes. Within MeDIP-Seq libraries, methylated cytosines can be found in both double-stranded (symmetric) and single-stranded (asymmetric) genomic contexts. While symmetric CG methylation has been relatively well-studied, asymmetric methylation in any dinucleotide context has received less attention. Importantly, no currently available software for processing MeDIP-Seq reads is able to resolve these strand-specific DNA methylation signals. Here we introduce DISMISS, a new software package that detects strand-associated DNA methylation from existing MeDIP-Seq analyses.ResultsUsing MeDIP-Seq datasets derived from Apis mellifera (honeybee), an invertebrate species that contains more asymmetric- than symmetric- DNA methylation, we demonstrate that DISMISS can identify strand-specific DNA methylation signals with similar accuracy as bisulfite sequencing (BS-Seq; single nucleotide resolution methodology). Specifically, DISMISS is able to confidently predict where DNA methylation predominates (plus or minus DNA strands - asymmetric DNA methylation; plus and minus DNA stands - symmetric DNA methylation) in MeDIP-Seq datasets derived from A. mellifera samples. When compared to DNA methylation data derived from BS-Seq analysis of A. mellifera worker larva, DISMISS-mediated identification of strand-specific methylated cytosines is 80 % accurate. Furthermore, DISMISS can correctly (p <0.0001) detect the origin (sense vs antisense DNA strands) of DNA methylation at splice site junctions in A. mellifera MeDIP-Seq datasets with a precision close to BS-Seq analysis. Finally, DISMISS-mediated identification of DNA methylation signals associated with upstream, exonic, intronic and downstream genomic loci from A. mellifera MeDIP-Seq datasets outperforms MACS2 (Model-based Analysis of ChIP-Seq2; a commonly used MeDIP-Seq analysis software) and closely approaches the results achieved by BS-Seq.ConclusionsWhile asymmetric DNA methylation is increasingly being found in growing numbers of eukaryotic species and is the predominant pattern observed in some invertebrate genomes, it has been difficult to detect in MeDIP-Seq datasets using existing software. DISMISS now enables more sensitive examinations of MeDIP-Seq datasets and will be especially useful for the study of genomes containing either low levels of DNA methylation or for genomes containing relatively high amounts of asymmetric methylation.

Project description:DNA CpG methylation is a widespread epigenetic mark in high eukaryotes including mammals. DNA methylation plays key roles in diverse biological processes such as X chromosome inactivation, transposable element repression, genomic imprinting, and control of gene expression. Recent advancements in sequencing-based DNA methylation profiling methods provide an unprecedented opportunity to measure DNA methylation in a genome-wide fashion, making it possible to comprehensively investigate the role of DNA methylation. Several methods have been developed, such as Whole Genome Bisulfite Sequencing (WGBS), Reduced Representation Bisulfite Sequencing (RRBS), and enrichment-based methods including Methylation Dependent ImmunoPrecipitation followed by sequencing (MeDIP-seq), methyl-CpG binding domain (MBD) protein-enriched genome sequencing (MBD-seq), methyltransferase-directed Transfer of Activated Groups followed by sequencing (mTAG), and Methylation-sensitive Restriction Enzyme digestion followed by sequencing (MRE-seq). These methods differ by their genomic CpG coverage, resolution, quantitative accuracy, cost, and software for analyzing the data. Among these, WGBS is considered the gold standard. However, it is still a cost-prohibitive technology for a typical laboratory due to the required sequencing depth. We found that by integrating two enrichment-based methods that are complementary in nature (i.e., MeDIP-seq and MRE-seq), we can significantly increase the efficiency of whole DNA methylome profiling. By using two recently developed computational algorithms (i.e., M&M and methylCRF), the combination of MeDIP-seq and MRE-seq produces genome-wide CpG methylation measurement at high coverage and high resolution, and robust predictions of differentially methylated regions. Thus, the combination of the two enrichment-based methods provides a cost-effective alternative to WGBS. In this article we describe both the experimental protocols for performing MeDIP-seq and MRE-seq, and the computational protocols for running M&M and methylCRF.

Project description:BackgroundClostridiales and Bacteroidales are uniquely adapted to the gut environment and have co-evolved with their hosts resulting in convergent microbiome patterns within mammalian species. As a result, members of Clostridiales and Bacteroidales are particularly suitable for identifying sources of fecal contamination in environmental samples. However, a comprehensive evaluation of their predictive power and development of computational approaches is lacking. Given the global public health concern for waterborne disease, accurate identification of fecal pollution sources is essential for effective risk assessment and management. Here, we use random forest algorithm and 16S rRNA gene amplicon sequences assigned to Clostridiales and Bacteroidales to identify common fecal pollution sources. We benchmarked the accuracy, consistency, and sensitivity of our classification approach using fecal, environmental, and artificial in silico generated samples.ResultsClostridiales and Bacteroidales classifiers were composed mainly of sequences that displayed differential distributions (host-preferred) among sewage, cow, deer, pig, cat, and dog sources. Each classifier correctly identified human and individual animal sources in approximately 90% of the fecal and environmental samples tested. Misclassifications resulted mostly from false-positive identification of cat and dog fecal signatures in host animals not used to build the classifiers, suggesting characterization of additional animals would improve accuracy. Random forest predictions were highly reproducible, reflecting the consistency of the bacterial signatures within each of the animal and sewage sources. Using in silico generated samples, we could detect fecal bacterial signatures when the source dataset accounted for as little as ~ 0.5% of the assemblage, with ~ 0.04% of the sequences matching the classifiers. Finally, we developed a proxy to estimate proportions among sources, which allowed us to determine which sources contribute the most to observed fecal pollution.ConclusionRandom forest classification with 16S rRNA gene amplicons offers a rapid, sensitive, and accurate solution for identifying host microbial signatures to detect human and animal fecal contamination in environmental samples.

Dataset Information

MethRaFo: MeDIP-seq methylation estimate using a Random Forest Regressor.

Publications

MethRaFo: MeDIP-seq methylation estimate using a Random Forest Regressor.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets