MS MS validation
Tandem mass spectrometry or MS/MS is an important technique in analytical Mass spectrometry. Processing MS/MS data for a large batch of samples can be a time-consuming task. We have added some support for processing such data in El-MAVEN.
Data acquired using the following methods are currently supported in El-MAVEN: Multiple Reaction Monitoring, Parallel Reaction Monitoring (or Full-Scan MS2) and Data Dependant MS2
- Text search option using precursor and product m/z
- Targeted Peak Picking for MS/MS
- Manual annotation of SRM transition
In order to ensure correct reading and processing of MS2 sample files in El-MAVEN, MS spectra and chromatograms from El-MAVEN were validated using Skyline by ProteoWizard.
Dataset: Full-scan MS2 sample file
| Fragment | Skyline | ElMaven |
|---|---|---|
| #1 |  |
 |
| #2 |  |
 |
| #3 |  |
 |
| #4 |  |
 |
| #5 |  |
 |
Dataset 1: Multiple Reaction Monitoring
| Fragment | Skyline | El-MAVEN |
|---|---|---|
| #1 |  |
 |
| #2 |  |
 |
| #3 |  |
 |
| #4 |  |
 |
Multiple Transition cases: The same precursor/product pairs can be tracked in multiple runs, for example, in cases where two metabolites have common fragments like Pyruvate 89/89 and Lactate 89/89. El-MAVEN lists these runs separately in the SRM List widget. Following are some examples:
| Fragment | Skyline | El-MAVEN | Comments |
|---|---|---|---|
| Lactate Fragment #1 |  |
Compound widget:  Both runs for this fragment have been annotated as Lactate in the SRM widget. Transition 1: Transition 2: 
|
Skyline EIC matches Transition 2 from the SRM widget in El-MAVEN |
| Pyruvate Fragment #1 | EIC is same as that of Lactate for this fragment. Different peak is selected according to the expected retention time. 
|
Compound widget:  Zoomed in:  SRM widget: No transitions mapped to Pyruvate |
Unable to find record of another transition in Skyline. The choppy peaks in El-MAVEN are due to merging of the two transitions annotated as Lactate |
Text search/Compound widget mismatch: Using the text search feature for MRM data can result in choppy peaks at times as compared to those from the SRM/compound widget. Following are some examples:
| Fragment | Skyline | El-Maven Text Search | El-Maven Compound widget |
|---|---|---|---|
| #1 |  |
 |
 |
| #2 |  |
 |
 |
Dataset 2: Full-scan MS2
| Fragment | Skyline | El-MAVEN Text Search | El-MAVEN Compound widget | Comments |
|---|---|---|---|---|
| #1 |  |
 |
 |
All three EICs match exactly |
| #2 |  |
 |
|
All three EICs match exactly |
| #3 |  |
 |
 |
EIC from the compound widget is different as one of the runs has been annotated as this fragment which pulls up EIC for the whole run |
- Text search and compound/SRM widget EIC mismatch for MRM data: EIC obtained from the compound widget and Skyline are the same. The data points on the plot are same for all three. The error in text search is due to different processing. This has been filed as issue number #487 Status: Unresolved
- Text search and compound widget EIC mismatch for PRM data: EIC from Text search and Skyline are the same. The error in compound widget happens due to linking of SRM ID to the compound widget. This pulls up the whole run instead of the particular fragment. This is a known issue #442 Status: Unresolved
- Merged peaks in case of multiple transitions: This is being handled as part of issue number #405 Status: In progress
- Mass spectra displays wrong product M/z value in some cases: The prodMz field in the title of spectra widget displays the base peak m/z value which may or may not be the same as the fragment m/z. This will be resolved as part of #396 Status: Unresolved
- Peak table displays NA in Expected m/z column for MRM data: This has been filed as issue number #493
Mass spectra validation against Skyline passed. This indicates correct parsing of MS/MS data in El-MAVEN. EIC validation against Skyline highlighted some known and unknown issues. We recommend using Text search for PRM data and Compound/SRM widget for MRM data processing until the reported issues have been fixed.