Home  •  Project  •  Resources  •  Publications  •  Help/FAQ  •  Contact


  1. How many tumors does database/website contain?

    Please go to the "home" page, and see the "Sample Statistics" section. The current release contains total of 232 samples.

  2. How can I view my favorite gene?

    In the navigation toolbar, you can either type the gene symbol (i.e. BRCA1) or the genomic coordinates (i.e. chr17:38,449,840-38,530,994), and then click the "Refresh View" button.

  3. How can I download data from this website?

    You can download data that covers a small region. Look for a button "Download 1M bp Region" at bottom-right of the genomic view page. Due to the large amount of data, we are unable to provide a download link for all the data at this time. We will release all data from the "Resources" section later. To download data, first users need to select regions by typing the gene symbol or region coordinates (see point 2). Second, users can choose a methylation intensity track (breast samples only, endometrial samples only, all samples or the summary information) and then click the "Refresh View" button. Finally, after all the tracks are visualized (i.e. loaded), users can click the "Download 1M bp Region" button to download the data. Each sample will be matched to one bed file. The region should be less than 1 M bp, otherwise only the 1st 1 M bp will be downloaded. The file is in bed format (sample1.bed).

  4. What's the input and output of differentially methylated region (DMR) function?


    Treat and control samples, user can choose samples by clicking the check box in the browser page.

    Normalization method: two normalization methods were included: linear normalization and quantile normalization. Default method is linear normalization.

    Maximum reads number: the maximum threshold for methylation intensity (for 100 bp bin size). The methylation level larger than the threshold will be removed. The default value is 100.

    Minimum reads number: most of the genomic regions have no or very little methylations. User can use this parameter to screen out the region with abundant methylations for DMR detection. The default value is 0.3.

    P-value threshold: set the threshold of the p-value of the statistical test. The default value is 0.01.

    Stat method: the statistical methods used for the DMR detection: t-test, Wilcoxon test and Pearson correlation coefficient are the choices.

    Regionlength: the window size of the specific region that was used for the DMR detection. This number should be larger than 100, which is the resolution of the methylation profiles track. The selected region is scanned by this sliding window size, and moved by a step defined below. The default length is 1000.

    Regionstep: the sliding step of the window. This number should be larger than 100, which is the resolution of the methylation profiles track. The default step is 500.

    Output (sample2.txt):

    Contains chromosome ID, start of the region, end of the region, p-value, methylation difference (average reads number difference in DMR region between treated and control sample(s)), methylation ratio (methylation difference divided by average reads number of control sample(s)).

  5. Upload customized tracks

    Users can upload their own data to the website by clicking the "Choose File" button, and then click the "Refresh View" button. The file must be in bed format, the same as "sample1.bed".

  6. Gene set organization of the Gene centric view

    Users can visualize a set of genes predefined in the database. The primary gene set names and their sources are listed in the following table. The "Filter Search" option allows a user to find all gene sets, except those among the "Correlated Genes", which contain the letters in the filter. The Gene Symbol option allows a user to search for a gene among all the 8 primary categories of gene sets.

    Gene Set Name Description Source

    Chromosomal Genes with a given chromosomal location MSigDB
    Gene Ontology Gene sets derived from gene ontology terms in all three GO categories MSigDB
    Perturbation sets Gene sets obtained from chemical and genetic perturbation MSigDB
    Biological Pathways Genes derived from various pathway systems MSigDB,
    Reactome, KEGG,
    NCI Nature, BioCarta,
    and HumanCyc
    microRNAs Genes that regulated by miRNAs MSigDB
    Transcription Factors Genes that regulated by transcription factors MSigDB
    Cancer gene neighborhood Genes that associated with 380 cancer genes MSigDB
    Correlated genes Genes that are correlated based on methylation status of the CMS 191 tumors