How to build spectral library using rawdata at small molecule interface whdgka75  2023-05-06 00:16
 

I want to build a spectral library through metabolome raw file, how can I build it?

I'm trying to next follow

  1. enter molecule interface
  2. menu tab -> settings -> molecule settings
  3. library tab -> build button
  4. input Name & output path -> Next
  5. Add files -> input raw data

but i'm met the "The following files are not valid library input files ..." error.

input the raw file is downloaded from metabolight database, study name is MTBLS3916.

also, not only those files but also the other raw files failed to build the spectral library.

there are various file extensions such as ".raw", ".dat", and ".wiff" ...

please help me..

 
 
Nick Shulman responded:  2023-05-06 09:39
I am not sure that I understand what you are trying to do but you probably do not want to build a spectral library.

The usual thing that people do with raw files in Skyline is extract chromatograms from them. In order to extract chromatograms you would first need to tell Skyline which molecules you are interested in, and then you would use the "File > Import > Results" menu item to tell Skyline to extract chromatograms from your .raw files.

If you want to learn about how to tell Skyline about the molecules that you are interested in, it might be helpful to look at the Small Molecule Targets tutorial:
https://skyline.ms/wiki/home/software/Skyline/page.view?name=tutorial_small_molecule

If you send us your files we might be able to give you better advice about what Skyline can do with those files.
You can upload files here:
https://skyline.ms/files.url

Alternatively, if you downloaded your files from some place on the internet you could tell us where you downloaded them from and we could take a look.
-- Nick
 
Brian Pratt responded:  2023-05-06 16:00
It looks like the file m_MTBLS3916_LC-MS_alternating_reverse-phase_metabolite_profiling_v2_maf.tsv would be a suitable transition list for this.
 
whdgka75 responded:  2023-05-07 21:59
I try to build a spectral library using tsv file, but met an error message

I attached an error message that occurs when I build the spectral library.

It needs to modify column names for building the spectral library, I think.

I have uploaded metabolome raw file which want to build the spectral library.

Please help me...
 
Brian Pratt responded:  2023-05-08 08:18
It looks like you're trying to use the "File > Import > Assay Library" function, which is not the workflow you're looking for. You don't have a library to import, you're trying to build a library based on a transition list and some raw data.

Start with "File > Import > Transition List". Do have a look at the tutorial Nick mentioned.

It's not clear to me, though, which columns in that transition list define the fragment m/z values if any, so this file may still not be quite what you need. Hopefully somewhere in the MTBLS3916 data set there exists some kind of justification for the provided molecule IDs - a search results file of some kind. Ideally you would just get a .MSP file from that site and load that into Skyline, but I don't know if that's provided or not.

It is important to understand that a raw file by itself is not a basis for a spectral library since it contains no identifications. You need the precursor and fragment m/z values. Skyline will use those to extract chromatograms in the raw data, and can then create your spectral library.
 
whdgka75 responded:  2023-05-09 18:05
I am trying to use building a spectral library file using raw data(ex:MTBLS3916).

Previously, Mr. Brian Pratt, you said that he seemed to have to use the tsv file, so I tried to build it with that file. (settings - molecule settings - library tab - build button and open tsv file)

But I got an error like the first image below (file name : error_msg1).

So I tried to open the tsv file in skyline. (file - import - transition list)

Then the second error like the second image below occurred. (file name : error_msg2)

If you look at the image, it seems that it does not recognize the required column value.

I thought that spectral library build could not be done because of this problem.

tsv file, like the third image, "https://skyline.ms/project/home/support/file%20sharing/begin.view?" Uploaded the MTBLS3916 data file to . (file name : img3)

Could you please give me an answer regarding this?

I am aiming to build a spectral library using metabolme raw data.
 
Brian Pratt responded:  2023-05-10 08:30
>> If you look at the image, it seems that it does not recognize the required column value.

You can use those controls at the top of each column to let Skyline know what each column means. There are no standards for column names, so this flexibility is important - and is covered in the tutorial Nick mentioned. For example, "mass_to_charge" would be set to "Precursor m/z", "metabolite_identifier" to "Molecule Name", "chemical_formula" to "Molecular Formula", etc.

The charge column is a problem - "positive" and "negative" isn't really very useful since it gives no hint as to the actual ionizing adduct. If you go into a text editor and change that .tsv file so that "positive" is now "M+H" and "negative" is "M-H", you get farther.

But finally once you have all the columns identified, Skyline reports that many of the declared m/z values make no sense with the declared chemical formulas. You could ignore the chemical formula column and proceed, but that doesn't sound like a very good idea.

And, ultimately, there is no fragment information provided so this isn't a good basis for a spectral library anyway. Skyline wouldn't know what to look for in the MS2 data

I'm not sure what to tell you beyond this.. It doesn't seem like this data set provides much justification for its purported identifications.