findmzROI: link information here with output from findPeaks.centWave

August 19, 2016, 02:53:10 PM

Hi folks,

I would like to compare the results from findmzROI with the output from findPeaks.centWave.

In my dataset, I know we have peaks that are not being resolved with our LC method. I would like to know how many different m/z values this is. Here, by not resolved, I mean I have many isomers that span multiple minutes in our LC run. I am most of the way through this thanks to some of the code and responses on this forum. However, I have found some undocumented output from findPeaks.centWave and am hoping that someone knows a little more about the behind the scenes steps to help me understand the output. Here's what I have done:

1. Find the list of ROIs using findmzROI as described by HP Benton (available here)
2. Run findPeaks.Centwave using my parameters of choice and verbose.columns = TRUE.

After step 1, I have a list of features (mz, mzmin,mzmax, scmin, scmax, length, intensity).
After step 2, I have everything described in the help, and a few additional columns: lmin, lmax, sample.

What I would like to do is to directly tie my results from step 1 with the shorter list in step #2. However, I don't see any obvious index/marker that does this. I could search with some narrow m/z and retention time window, but I would rather know which ROI corresponds to which feature without setting some bounds on search parameters.

I hoped that 'f' (described as the 'region number of m/z ROI where peak was localised') would be helpful, but I am not sure how to use this information.

Thanks in advance for thoughts on this.

From sessionInfo()
R version 3.0.3 (2014-03-06)
XCMS_1.38.0