Shimadzu Library file not loading Alan Barnes  2019-04-08
 

Dear Skyline Team,

I am having trouble importing a Shimadzu Library file. I have followed the following path is this correct?
Settings > Peptide Settings > Library > Edit LIst > Add > Browse

I have found mlb file format is not read.
https://skyline.ms/wiki/home/software/Skyline/page.view?name=building_spectral_libraries&_docid=wiki%3A58131659-9e41-1031-aa4c-da2025825caa

 
 
Brendan MacLean responded:  2019-04-08

Hi Alan,
You are the first person to mention this new format to us. Is this from a Shimadzu DDA search tool created for the new Q-TOF instrument?

It appears to be a binary format file. Is it a standard binary format like SQLite, or a proprietary custom format defined by Shimadzu? If the latter, we would likely need a new DLL from Shimadzu to allow us to read it.

We are happy to work on this with you, but this is the first we have heard of it, and we will definitely need more information to get started on supporting it.

Thanks for bringing this to our attention.

--Brendan

 
Alan Barnes responded:  2019-04-09

Hi Brendan,

The example I attached was from the QTOF from a peptide digest with just a few spectra registered. I sent you and Brian a PM, I am happy to provide more data examples.

Best regards,
Alan

 
Brian Pratt responded:  2019-04-15

Hi Alan,

The problem is just that you need to convert the .mlb file to .blib format before adding it. This is true of most spectral library formats that Skyline deals with. I can see where expecting the conversion step to happen automatically is a reasonable user expectation, but it's just not how it works at the moment.

That said, this particular example .mlb is rather strange compared to what we've seen before - all molecules have the same name and adduct, and no chemical formula. All that differs is RT and precursor m/z. As we use molecule name and adduct as the library locator key, there won't be much we can do with this file - from Skyline's point of view its contents are inherently in conflict with each other.

It does appear that we're not properly filtering out the MS1 data, which I'll see to, but the other issues remain. I'm happy to work with your engineers to find a solution.

Best Regards,

Brian