Support for viewing methylation with Illumina 5-base sequencing

[aferriz]

Hi,

Is there a way to visualise Illumina 5-base methylation data in IGV, or are there plans to include support for it?
Currently, when using the available methylation viewing modes, the bar heights show the VAF of individual CpG positions rather than the combined methylation signal, which makes comparison with ONT data difficult.

I've attached a screenshot showing Illumina 5-base data (top) and ONT data (below).

Another picture with Illumina 5-base (coloured by bisulphite mode CG) on top and ONT data below:

Any help would be appreciated.
Thank you!
Best,
Alvaro

[jrobinso]

I'm not familar with Illumina 5-base sequencing, and I don't know what options you are using for coloring. However in the upcoming release there is an option to set the cytosine context in preferences. You can try this with the development snapshot build available here: https://igv.org/doc/desktop/#DownloadSnapshot/. Select "View > Preferences > Base Mods" and scroll to the bottom. Set the context to CG and see if it does what you are looking for.

[aferriz]

Hi @jrobinso, thanks for the pointer, I tried the snapshot version, but it behaves much the same as the current options.

It would be really helpful if IGV could add a proper methylation visualization mode for Illumina 5-base sequencing data, similar to what's supported for ONT. With adoption of this technology growing, I imagine more users will be looking for this. Appreciate your help!

[jrobinso]

@aferriz Did you set the context preference to CG in the snapshot version? Could you make a screenshot of your "Base Mods" setting page in the snapshot?

IGV doesn't distinguish ONT vs PacBio vs Illumina. It supports the standards for base mod tagging in the bam/cram format. As long as the data is encoded correctly it should not matter which platform produced it. However there are a several options for viewing the data, 1-color vs 2-color mode, and in the snapshot no context (default) vs defined context (set through preferences).

To make further progress I would need some example data.