Skip to main content
Topic: Clarify gap filling by compulsion parameter (Read 614 times) previous topic - next topic

Clarify gap filling by compulsion parameter

I'm looking for clarification on the gap filling by compulsion parameter under the alignment options.

My data (see attached) show a significant effect when this parameter is enabled (top histogram) vs disabled (bottom histogram). My dataset has 60 samples, and I receive a total of approximately 18,000 features before filtering steps.

Section 2.3.6 of the official docs https://mtbinfo-team.github.io/mtbinfo.github.io/MS-DIAL/tutorial.html  gives the following:

Quote
If you check this, the peak recognition is performed by the average peak width of samples having the metabolite feature even though no local maximum is observed in the chromatogram. This is validated by default.

  • The "average peak width of samples" is the average FWHM of all detected peaks across all samples within one feature. Correct?
  • How does having an average FWHM enable finding peaks with no local maxima?
  • How does this parameter cause such a large difference in the number of non-zero values for each feature?
  • Based on the parameter "by compulsion" it makes sense that there should be no zero values at all with this parameter enabled, but I still have some (top histogram). Why is that?

An explanation of the parameter algorithm would answer these questions.

Also, reading this prior post did not answer my questions. 

Re: Clarify gap filling by compulsion parameter

Reply #1
Hi. I untick the "gap filling by compulsion", but the result file is the same with the one with gap filling. I am so confused how I can cancel the gap filling.