AnsweredAssumed Answered

Trying to Understand Library Match Factors, MH B.08

Question asked by trifecta on Nov 12, 2018
Latest reply on Nov 15, 2018 by tom_barrett

Like • Show 0 Likes0
Comment • 4

I'm hoping someone can help me understand the library match factors from Quant MH B.08. The attached pdf shows 3 examples. In the first example, the match factor or match score of 89 makes sense to me. Most of the masses are present and there are no extra masses in the unknown spectrum (top). The middle overlay shows a match of 41.5. This doesn't look right to me as only 1 of 4-6 major masses is present, but it is one of the library masses. The last example really has me wondering what is going on. I agree the unknown spectrum is only one mass but the match of 30 is strange. None of the library masses are present. I'm open to ideas. Thank you.

tom_barrett

Nov 14, 2018 5:21 PM

Correct Answer

Hi Trifecta,

On the surface the last example is in enigma but I think that I may know the answer. My theory is that the sample spectra actually has some very, very small ions that we don't see due to scaling issues. One or more of the ions align with the library spectrum and we get a non zero LMS. I don't have the data to test this but if you will expand the y-axis you can check for this.

Tom

See the reply in context

Attachments

Match_Factors_1.pdf208.7 KB

No one else had this question

Outcomes

tom_barrett

Nov 14, 2018 5:21 PM

Hi Trifecta,

Tom

Actions

trifecta @ tom_barrett on Nov 15, 2018 12:19 PM

Hello Tom,

Thank you for your suggestion. I did zoom the y-axis and found one mass peak, 57, at an abundance of 0.8. Picture attached. This compares to the base peak mass 74 in the unknown of 1000 abundance. I could not find a way to directly access NIST MSSearch from MH Quant. Therefore I created a spectrum manually in NIST MSSearch, with these 2 mass-abundance pairs. I also created a NIST library of one entry, the compound of interest, using the NIST spectrum. I searched the unknown spectrum against this single spectrum library. The forward match factor was 0 (zero) and the reverse match factor was 70. This seemed a bit generous to me, but 57 is the base peak and if it's the only mass considered, well OK.

Can Agilent provide a document that shows the equation for the Agilent Library Search Match Factor ?

Also, can Agilent provide a document that describes whether reverse of forward search is used, and if both are used what are the weighting factors ?

Thanks for your help.

Trifecta

Actions

trifecta @ trifecta on Nov 15, 2018 12:21 PM

Actions

tom_barrett

@ trifecta on Nov 15, 2018 12:51 PM

Hi Trifecta,

Good to know I was right about the tiny ions. I have contacted the R&D team to see what sort of documentation we have but can tell you a few things now. The weighting in MH is 70 forward & 30 reverse. In Unknowns Analysis B.09.00 this is exposed and user adjustable in the method.

Will post more when I know more.

Tom

Actions

4 Replies

tom_barrett Nov 14, 2018 5:21 PM
Unmark Correct
Correct Answer
Hi Trifecta,

On the surface the last example is in enigma but I think that I may know the answer. My theory is that the sample spectra actually has some very, very small ions that we don't see due to scaling issues. One or more of the ions align with the library spectrum and we get a non zero LMS. I don't have the data to test this but if you will expand the y-axis you can check for this.

Tom
Like • Show 1 Like1
Actions
- trifecta @ tom_barrett on Nov 15, 2018 12:19 PM
  Mark Correct
  Correct Answer
  Hello Tom,
  Thank you for your suggestion. I did zoom the y-axis and found one mass peak, 57, at an abundance of 0.8. Picture attached. This compares to the base peak mass 74 in the unknown of 1000 abundance. I could not find a way to directly access NIST MSSearch from MH Quant. Therefore I created a spectrum manually in NIST MSSearch, with these 2 mass-abundance pairs. I also created a NIST library of one entry, the compound of interest, using the NIST spectrum. I searched the unknown spectrum against this single spectrum library. The forward match factor was 0 (zero) and the reverse match factor was 70. This seemed a bit generous to me, but 57 is the base peak and if it's the only mass considered, well OK.
  Can Agilent provide a document that shows the equation for the Agilent Library Search Match Factor ?
  Also, can Agilent provide a document that describes whether reverse of forward search is used, and if both are used what are the weighting factors ?
  Thanks for your help.
  Trifecta
  Like • Show 0 Likes0
  Actions
  - trifecta @ trifecta on Nov 15, 2018 12:21 PM
    Mark Correct
    Correct Answer
    Like • Show 0 Likes0
    Actions
  - tom_barrett @ trifecta on Nov 15, 2018 12:51 PM
    Mark Correct
    Correct Answer
    Hi Trifecta,
    
    Good to know I was right about the tiny ions. I have contacted the R&D team to see what sort of documentation we have but can tell you a few things now. The weighting in MH is 70 forward & 30 reverse. In Unknowns Analysis B.09.00 this is exposed and user adjustable in the method.
    Will post more when I know more.
    Tom
    Like • Show 0 Likes0
    Actions

Trying to Understand Library Match Factors, MH B.08

Attachments

Outcomes

Related Content

Recommended Content