MSP Spectral Libraries for Small Molecules

support
MSP Spectral Libraries for Small Molecules Christina lucas  2020-04-22
 

Hi there,

i am trying to Import an MSP library to Skyline.

First attempt was an Test Library created in ThermoFishers mzVault exported as MSP (attached). Which was not working, no element was shown.
Second attempt, as Thermo might have some Export issues on that was to download the HMDB (or any other file in MSP) from https://mona.fiehnlab.ucdavis.edu/downloads

Still I get no content in my library.

So far I could not find anything in Forum thats giving me a hint how to open them correctly.
I am using daily 20.1.1.83 on a Win 10.

Best,

Christina

 
 
Brian Pratt responded:  2020-04-22

Hi Christina,

Thanks for the example file. MSP is another one of those formats that's an ad hoc "standard" with different variants popping up now and then.

The problem in this case is that there's no ionization information ("Precursor_type:") so Skyline ignores the input. It shouldn't ignore it silently, though, I'll see about improving that.

There are, although, ionization hints:
Ionization: ESI
IonMode: positive
PrecursorMz: 204.12315
Formula: C9 H17 N O4

With these taken together we can deduce that ionization is in fact [M+H], but that's logic I'll have to add. Thanks for bringing it to our attention. Clearly we want to be compatible with the likes of MassBank of North America.

Thanks for using the Skyline support board!

Brian Pratt

 
Brian Pratt responded:  2020-04-22

This one's a little tricky:

Name: C6:OH b
PrecursorMz: 258.17001
Ionization: ESI
IonMode: positive
Formula: C13 H25 N O5

I imagine that's [M-H2O+H]? Pretty annoying that they don't just say so. Are there other common adducts you think we should be able to infer?

Thanks

Brian

 
Christina lucas responded:  2020-04-23

Hi Brian,

thanks for the fast reply.

I checked in the mzVault Library Editor (Thermo) whether there is an opportunity to put in the Ionisation Adduct.
It does not seem so.

You can add SMILES and InChi, CAS and so on...But that would not help any further.
I did not take a look inside the structure of the MSP HMBD database from MONA if they add the adduct.
But I would assume No, because on the provided Libraries on the HMDB Website (XML and TXT Format) do not contain an information about this.
I took a look at MASSBANK, the include the precurser type.

For adducts maybe MeOH (CH3OH) or Acetonitrile (C2H3N) adducts could occur, but I would not call it common.
Another questions that came up while thinking about this, how about multiple charges ? Does Skyline realize it when checking the formula and the mass ? Here (M+2H ... ) might play a role...
The loss of water is of course an issue.

Thanks for your help,

Christina

 
Brian Pratt responded:  2020-04-23

It's so interesting that this information gets discarded so often. It seems pretty fundamental?

We do check more than just +1, no problem there. I'll extend the code to check for water loss, and the more common ionization types as seen in the table at https://fiehnlab.ucdavis.edu/staff/kind/Metabolomics/MS-Adduct-Calculator/

Thanks for helping improve Skyline!

Brian

 
Christina lucas responded:  2020-04-28

Hi Brian,

sorry for the delay.

Well, maybe its assumend everything inside those libraries is M+H / M-H, if not stated otherwise.

When do you think this could be implented?

Thanks for getting back so quickly all the time!

Best,

Christina

 
Brian Pratt responded:  2020-04-28

Hi Christina,

Well, maybe its assumed everything inside those libraries is M+H / M-H, if not stated otherwise.

Unfortunately as we've seen there may also be implied water loss, so they really haven't stated otherwise. This seems like critical information, very strange that it should be left out.

When do you think this could be implemented?

Already done. It will be in the next Skyline-Daily, which I think will be as soon as next week.

Thanks for getting back so quickly all the time!

It's what we do! Please just make sure your mass spec manufacturer knows how much you appreciate their support of the Skyline project.

Best,

Brian