Question: Visualizing 450 K Methylation Array Data In Igv
gravatar for kanwarjag
6.5 years ago by
United States
kanwarjag1.0k wrote:

I have a 450 K infinum DNA methylation array data with methylation b values. I would like to plot it in any genome browser prefer is IGV. I can convert generate coordinates bed file and then upload to genome browser- IGV/ UCSC. My question is can I vary the histogram as per level of beta values? Is there any other nice way of visualizing 450 methylation data in genome browser?


array methylation igv • 4.9k views
ADD COMMENTlink modified 6.5 years ago by Charles Warden7.6k • written 6.5 years ago by kanwarjag1.0k
gravatar for dario.garvan
6.5 years ago by
dario.garvan460 wrote:

BED format is not the right format. You need to output the data in bedGraph format.

ADD COMMENTlink written 6.5 years ago by dario.garvan460

I also agree that bedgraph is the format that would be more helpful for your interests. you can define start and end chromosome positions, plus a value that usually represents coverage, but that you can use it to represent the methylation b values instead. you can still gzip the file, and IGV will still be able to deal with it. a similar approach (reducing file size to improve data storage and make data access faster) would be to transform that bedgraph file to bigwig format using UCSC's bedGraphToBigWig (find it here), which is an indexed binary format that works great on IGV. plus you can apply colors to this track on IGV depending on thresholds, which is useful to detect high and low peaks for instance.

ADD REPLYlink modified 6.5 years ago • written 6.5 years ago by Jorge Amigo11k
gravatar for Charles Warden
6.5 years ago by
Charles Warden7.6k
Duarte, CA
Charles Warden7.6k wrote:

COHCAP (using the "Average per Site" workflow) will create .wig files for 450k methylation values (per site, per group as well as for delta-beta values, if you are comparing two groups), which can be visualized using IGV, UCSC, etc.

Outside of the genome browser, COHCAP can also create box-plots and correlation plots, if you have paired gene expression data (using the "Average per Island" workflow).

I think this is a pretty good strategy, but of course I am biased ;)

ADD COMMENTlink written 6.5 years ago by Charles Warden7.6k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2115 users visited in the last hour