MMASS FRAGMENTATION SOFTWARE SOFTWARE

MAGMa+ is a parameter-optimized version of the original MAGMa software. CFM-ID can be used to assign fragments to spectra to rank the candidates, but also to predict MS/MS spectra from structures alone. CFM-ID (competitive fragment modelling) employs a method for learning a generative model of collision-induced dissociation fragmentation. MetFragCL retrieves candidate structures and fragments them using a bond dissociation approach and those fragments are then compared to the product ions in a measured spectrum to determine which candidate best explains the measured compound by assigning it a score that is a function of the mass to charge ratio ( m/z), intensity and bond dissociation energy (BDE) of the matched peaks, while 5 of neutral loss rules account for rearrangements. We surveyed four different tools, all using different algorithms for in silico fragmentation, MetFragCL, CFM-ID, MAGMa+ and MS-FINDER.

MMASS FRAGMENTATION SOFTWARE SOFTWARE

For practical reasons, including the source code and model availability, error handling, batch processing capabilities and the ability to perform local database queries, we only covered in silico fragmentation software that was used for results submitted by the CASMI 2016 deadline. The Critical Assessment of Small Molecule Identification (CASMI) contest was founded in 2012 to help scientists with their compound identification methods by providing community challenges and competitions. The data for our investigation was obtained from the CASMI website ( ). A number of commercial software solutions such as MassFrontier (HighChem), MS-Fragmenter (ACD/Labs) or Molecular Structure Correlator (Agilent) are also available, but lack open access code or algorithm transparency. Such software tools include MetFrag, MIDAS, MAGMa, MAGMa+, MOLGENвЂ“MS/MS, CSI:FingerID, CFM-ID, FingerID, Input output kernel regression (IOKR) and the MS-Finder software. As an alternative strategy for compound annotation of known compounds in silico fragmentation software tools have been developed and are used to identify MS/MS spectra when the reference MS/MS spectrum is not available. Overall, tandem mass spectral databases cover less than one per cent of the compound space that is covered in Chemspider or PubChem with 50 to 90 million compounds, respectively. Such an approach yields hundreds to thousands of signals per study, most of them having unidentified structures even after comprehensive searches of existing mass spectral libraries such as NIST, MassBank, Metlin or MassBank of North America (MoNA). Many fields of research, from environmental analysis to forensics and biology, are moving towards hypothesis-generating screening approaches using liquid chromatography coupled with tandem mass spectrometry (LCвЂ“MS/MS). Matching MS/MS spectra against the MS/MS libraries without using any in silico tool yielded 60% correct hits, showing that the use of in silico methods is still important. We comprehensively analysed how outcomes of different tools could be combined and reached a final success rate of 93% for the training data, and 87% for the challenge data, using a combination of MAGMa+, CFM-ID and compound importance information along with MS/MS matching. We found that optimizing the use of metadata, weighting factors and the manner of combining different tools eventually defined the ultimate outcomes of each method. Performance of four publicly available in silico fragmentation algorithms (MetFragCL, CFM-ID, MAGMa+ and MS-FINDER) that participated in the 2016 CASMI challenge. The only way at present to address this discrepancy is to use in silico fragmentation software to identify unknown compounds by comparing and ranking theoretical MS/MS fragmentations from target structures to experimental tandem mass spectra (MS/MS). Without the true identity of these molecules it is impossible to draw conclusions about the biological mechanisms, pathway relationships and provenance of compounds. In mass spectrometry-based untargeted metabolomics, rarely more than 30% of the compounds are identified.

Mmass fragmentation software