How to Use DMineR: Prostate Cancer
There are three ways to find DMRs of interest:
- (1) Browse the DMR list by clicking on "DMR List" and then select a sub-set from the "DMR Classes" list on the left. Note that "Frequent 001" corresponds to "High Grade DMRs" and "Frequent 011" corresponds to "Shared DMRs" in the manuscript.
- (2) Enter a gene symbol in the search box on the top right to see all DMRs for which it is the nearest gene.
- And, (3) Enter a "dmrid" number from the DMR list published in the manuscript in the search box on the top right to directly view an exact DMR.
Each DMR has a profile page that can be viewed by clicking on the "Full Report" link from the DMR List. This page aggregates a variety of useful primary data, metadata, and linked data that we have found invaluable for interpreting the genomic and functional contexts of DMRs:
- The "Region count Boxplot" and "Region Overview" display the read counts from our primary DNA methylation sequencing data, which was obtained from the genome-wide technique MBD-isolated Genome Sequencing (MiGS).
- Next, "Gene Overlaps" displays all UCSC knownGene transcripts that overlap with the DMR.
- The "Feature Overlaps" shows ChIP-seq peaks and DNaseI hypersensitive sites from ENCODE that overlap with the DMR.
- Under "Omics Overlaps" a variety of data from other experiments is aggregated, as indicated by the "Omic Set" column. This includes gene expression and methylation microarray data from The Cancer Genome Atlas PRAD cohort. It also includes our own data (Ting Lab, Cleveland Clinic) from RNA-seq and MiGS data for a few prostate cancer cell lines. For frequent 001 DMRs that overlap with genes, there is also a link to PubMed queries that look for previous publications about the function of the gene relating to cancer.
Example DMRs
About the Data
Please see our manuscript (in preparation) for more information. Please cite this paper if DMineR or our primary data is useful to your work.