robinschmid
commented
2 months ago I actually don't know how to use this module. But just one question, most compounds in your blank will Not have a good ms2 spectrum because of relatively low intensities. Will this actually remove that many Features. I know that people usually just apply blank subtraction based on ms1
Basic information
What happened
I am currently working on identifying natural compounds from plant extracts.
I have conducted technical replicates three times, and to obtain more reliable data, I plan to align all these runs (three LC-MS data from plant extracts) and then proceed to remove blank peaks from them.
Therefore, I intend to compare the MS2 similarity between the blank and sample, and remove any cases where they exhibit similar fragment patterns.
I wanted to manually check for similarities, so I used the (feature list method—annotation—search spectra—)MS2 similarity search. However, there are a few unresolved questions I have.
https://mzmine.github.io/mzmine_documentation/module_docs/id_ms2_similarity/ms2-similarity-search.html Based on the above explanation, I understand that the MS2 similarity search calculates the similarity of MS2 spectra between the two feature lists and then outputs the results in a column of feature list 1.
and I used the following parameters:
And the results I got were incomprehensible.
I set the ppm to 5, but it matched ions that exceeded this value. Furthermore, the m/z 787.1004 ion is not a peak that exists in the blank.
What I am wondering is whether the MS2 similarity search compares feature 1 and 2 across different feature lists, or if it performs the search within the same feature list.
Even if the latter is correct, why are values that exceed the set parameters being matched?
Is there a way to manually compare the MS2 similarity between ions present in the blank and those present in the sample?
How do you utilize blank samples in your analyses? Or is it acceptable to proceed with ID without removing the blank?
If molecular networking is conducted with data that includes the blank, wouldn’t it be difficult to distinguish whether the observed compounds are derived from the blank or from the actual sample?
Thank you for taking the time to read my lengthy message. I look forward to your valuable responses.
Thank you. Sincerely, Jiwon
https://drive.google.com/file/d/1gpO9xw9i27s8LkrgXb8kRp0Yb1ewV2va/view?usp=sharing The attached project file contains only one experimental dataset, included to provide an example for my questions.