Increasing Read Usage in DMR

[maceyslota]

Hello,

I am wanting to see if it is possible to use more than the 300 read limit hard coded into the DMR command? Or, if there is a particular reason for the 300 read limit. I am looking for some potentially rare events and feel that 300 reads isn't capturing enough information. This is the error I get when I try to increase the max coverage to 300k reads:

calculated max coverage 300000 is greater than maximum allowed (300), setting to 300

Thanks!

[ArtRand]

Hello @maceyslota,

Let me look into it. The limit is due to the MAP-based p-value becoming unstable if you have very high coverages (as the p-value becomes zero). I'm sure I could add an option that removes the limit and disables the MAP-based p-value calculation - would that get you what you need? You would still get the score, effect, size, measurements of methylation etc.

[maceyslota]

Thanks for your response @ArtRand . I think the approach you described could be informative for the research question I am addressing. Thank you for offer to implement this change. Please let me know when you update it and how to use the options.

Do you think a different statistical test or model could be used to assess modification changes with higher coverages? If so, is this something that could be implemented?