-Part of the Imprinting track collection, this track summarizes allele-specific methylation (ASM) across +Part of the Imprinting track collection, this track summarizes allele-specific methylation (ASM) across thousands of human methylomes. A region where one allele is methylated and the other is not in many independent samples is a candidate imprinting control region.
Whole-genome bisulfite sequencing reads are long enough to cover several CpG sites at once, so the methylation states of neighboring sites on the same DNA molecule can be read together. When the reads at a locus fall into two clearly separated groups, one mostly methylated and one mostly unmethylated, that locus shows allele-specific methylation. This track shows, for every CpG site in the genome, the fraction of MethBase2 human methylomes in which the site was called part of an allelically methylated region (AMR). Values range from 0 to 1. High values mean that many independent samples show two differently methylated alleles at that site, which is the expected signature of a constitutive imprinted region. Lower but non-zero values point at loci where allele-specific methylation is restricted to certain tissues or is driven by sequence variation rather than by imprinting.
The score is a summary across samples, not a measurement in any single sample. A value of 0.5 does not mean that half of the DNA molecules are methylated; it means that about half of the surveyed methylomes had an AMR call covering that CpG.
The track is a signal graph with one value per CpG site. The default vertical range is 0 to 0.5, since even the strongest known imprinted regions rarely exceed that fraction across a set of samples this diverse. The range, the graph height, the smoothing window and the aggregation function can all be changed on the track configuration page. Positions between CpG sites carry no data and are drawn as gaps.
The per-sample methylation levels, hypomethylated regions and read coverage that this summary is derived from are published by the Smith lab as a track hub, which can be loaded from http://smithlab.usc.edu/trackdata/methylation/hub.txt or found in the list of public hubs.
MethBase2 collects publicly available whole-genome bisulfite sequencing runs from the NCBI Sequence Read Archive and reprocesses all of them through one uniform pipeline built on DNMTools. Reads are mapped with abismal, duplicates are removed, the bisulfite conversion rate is estimated, and per-site methylation levels are computed and symmetrized across the two strands of each CpG. A methylome enters the browsable collection only if its bisulfite conversion rate is at least 0.9 and at least 70% of CpG sites are covered, which in practice implies well above 1x average coverage. Allelically methylated regions are then called per methylome with the amrfinder command. It slides a window of 10 CpG sites along the genome and fits two models to the reads in each window, one in which both alleles share a single methylation state and one in which the two alleles have separate states. The window is called an AMR when the two-allele model fits significantly better, using a likelihood ratio test with a false discovery rate correction; windows need at least 4x mean coverage per CpG to be tested, and neighboring AMRs closer than 1 kb are merged. The statistical model is described in Fang et al. (2012). The score shown here is the number of methylomes whose AMR set covers a given CpG, divided by the number of methylomes tested.
The data file was downloaded from http://smithlab.usc.edu/methbase/data/common/hg38/hg38.asm.bw, the "common" section of the MethBase2 track hub at http://smithlab.usc.edu/trackdata/methylation/hub.txt. The file is a bigWig and was used as it is, without reformatting or recalculation. It carries a value at each of 29,401,795 CpG positions, on chr1 to chr22, chrX, chrY and chrM. The steps we ran are documented in the imprinting makeDoc.
The data can be explored interactively in table format with the Table Browser or the Data Integrator and exported from there to spreadsheet or tab-sep tables. From scripts, the data can be accessed through our API, track=methBaseAsm.
For automated download and analysis, the annotation is stored in a bigWig file that can be downloaded from our download server. The file for this track is called hg38.asm.bw. Individual regions or the whole annotation can be obtained using our tools bigWigToBedGraph or bigWigSummary, which can be compiled from the source code or downloaded as precompiled binaries for your system. Instructions for downloading source code and binaries can be found here. The tools can also be used to obtain values within a given range, e.g. bigWigToBedGraph http://hgdownload.soe.ucsc.edu/gbdb/hg38/imprinting/methBaseAsm/hg38.asm.bw -chrom=chr15 -start=24950000 -end=25200000 stdout
The original data can be downloaded from the MethBase2 track hub at http://smithlab.usc.edu/trackdata/methylation/hub.txt.
Thanks to Andrew D. Smith and the Smith lab at the University of Southern California for building MethBase2 and for making the allele-specific methylation summary available. The collection is maintained at https://github.com/smithlabcode/methbase.
Fang F, Hodges E, Molaro A, Dean M, Hannon GJ, Smith AD. Genomic landscape of human allele-specific DNA methylation. Proc Natl Acad Sci U S A. 2012 May 8;109(19):7332-7. PMID: 22523239; PMC: PMC3358917
Song Q, Decato B, Hong EE, Zhou M, Fang F, Qu J, Garvin T, Kessler M, Zhou J, Smith AD. A reference methylome database and analysis pipeline to facilitate integrative and comparative epigenomics. PLoS One. 2013;8(12):e81148. PMID: 24324667; PMC: PMC3855694