search

sneumann / xcms

This is the git repository matching the Bioconductor package xcms: LC/MS and GC/MS Data Analysis
Other
178 stars 81 forks source link

Re: questions about some data processing parameters #605

Closed JianFelixGuo closed 2 years ago

JianFelixGuo commented 2 years ago

Dear developer,

As a loyal user of XCMS, I usually use the default parameters for mzdiff = 0.01 and integrate = 1 in CentWave algorithm. I am wondering if I should change these 2 values based on certain criteria when I process different kinds of datasets. Do you have any experience of using different values for certain types of data?

It would be greatly appreaciated if you could kindly enlighten me on these questions. Thank you!

Best,

Jian

jorainer commented 2 years ago

I don't have any clear rules on that. I'm usually using integrate = 2 if the peaks have a long tail. With integrate = 1 centWave assumes more symmetric peaks and cuts the tail. For mzdiff I usually use mzdiff = 0.001 - but more out of a gut feeling (maybe your choice might be better).

JianFelixGuo commented 2 years ago

Thank you so much for the quick reply! I will try using your parameters.