Is there a way to run dmr pair with multiple replicates with a region BED file?

[lance0499]

Hi,

I have 5 control samples and 5 treated samples, and would like to run dmr analysis on only the CpG islands to compare the two groups.

However, the only option I can see for multiple replicates is at single base positions (dmr pair with multiple -a and -b). Using dmr multi will just do pairwise comparisons as well, which will not compare all replicates in a group-based setting.

Is there any way to be able to run DMR analysis with multiple replicates comparing two groups, while only focusing on specific regions like CpG islands?

Thanks a lot for the help!

[ArtRand]

Hello @lance0499,

Currently there isn't a way to perform a region-based pairwise comparison using multiple replicates where the replicates are combined in some way. I'm working on something, but it's not ready yet. What I recommend is to run dmr multi like you have then look at regions that rank highly in all treatment/normal comparisons.

[wietingj]

@ArtRand, since I have the same problem as described above, I was wondering if you were able to implement some way to use modkit dmr pair with a regions bed with multiple samples at once per group in the meantime?

Output (modkit v0.3.1):

Error! in order to perform multiple comparisons over regions use modkit dmr multi

Many thanks for your feedback!

[ArtRand]

@wietingj Others have asked for this functionality as well, it'll get into the next big release.

[lance0499]

Hello,

Is this feature added yet?

[ArtRand]

@lance0499

Not yet, should be in the next major release.

[Yang990-sys]

Hello, Is this feature added in this new version ?

[ArtRand]

Hello @Yang990-sys , @lance0499 , @wietingj

As of v0.4.0 when you use the same sample name in modkit dmr multi the samples will be combined.

I'm working on some other ways to tease apart multiple samples and allow for exploration of within and inter-sample variablility bt it's not quite ready. Let me know if the current version solves your use case.