search

RobertsLab / resources

https://robertslab.github.io/resources/
19 stars 11 forks source link

Decision on tiling analysis parameters #445

Closed yaaminiv closed 6 years ago

yaaminiv commented 6 years ago

Lab notebook entry

I played around with tiling window analysis in methylKit using three different sets of parameters:

  1. 100 bp window and step size
  2. 1000 bp window and step size (the methylKit default)
  3. 1000 bp window and 100 bp step size

I used the object I produced previously with mincov = 3.

Window Size (bp) Step Size (bp) Total Regions Number of Significantly Different DMRs
100 100 217538 162
1000 1000 104144 118
1000 100 104144 118

The first option provided the best separation between ambient and treatment samples in the PCA. The number of significantly different DMRs aren't that different between methods, so I thought it could be best to get as many DMRs as possible. Thoughts @sr320? I would like to finalize the method so I can match my DML and DMR lists with genes for the OAKL meeting/to update the methods in our manuscript.

sr320 commented 6 years ago

Sure - go with the 100bp windows

thanks Steven

Steven B. Roberts Kenneth K. Chew Endowed Professor School of Aquatic and Fishery Sciences University of Washington, Seattle WA sr320@uw.edu - 206.866.5141 robertslab.info - @sr320 On Oct 19, 2018, 4:55 PM -0700, Yaamini Venkataraman notifications@github.com, wrote:

Lab notebook entry I played around with tiling window analysis in methylKit using three different sets of parameters:

  1. 100 bp window and step size
  2. 1000 bp window and step size (the methylKit default)
  3. 1000 bp window and 100 bp step size

I used the object I produced previously with mincov = 3. Window Size (bp) Step Size (bp) Total Regions Number of Significantly Different DMRs 100 100 217538 162 1000 1000 104144 118 1000 100 104144 118 The first option provided the best separation between ambient and treatment samples in the PCA. The number of significantly different DMRs aren't that different between methods, so I thought it could be best to get as many DMRs as possible. Thoughts @sr320? I would like to finalize the method so I can match my DML and DMR lists with genes for the OAKL meeting/to update the methods in our manuscript. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.