The extensive chemical space for compounds that may be observed in real world sample analysis far outstrips the current size of available reference spectral libraries. This leaves many compounds unidentified by traditional spectral library matching approaches. Advances in spectral library searching attempting to provide putative identifications based on fragment similarity and other approaches have helped to address this gap but often do not provide enough information to establish an identification. Despite the large chemical space, there are many common substructures present that are the “building blocks” of molecules. Since a chemical structure will provide nearly identical fragments no matter what larger molecule it has come from, we can identify substructures in an unknown if we have a fragment spectra of them. A library with significant chemical diversity and MSn data will enable the identification of substructures in unknown chemical entities and allow for putative identification. The construction of such a library (mzCloud) has been an ongoing project and currently has curated reference MS/MS and MSn data on more than 19,000 standards with 3 million fragmentation spectra. For an unknown, the MSn spectra can be searched against the library to provide matches and the precursor of the matching reference spectra provides the chemical substructure information to the proposal for the unknown structure. We have applied this approach and the spectral library to the identification of ‘unknown’ compounds as a proof of principle. The compounds were not in mzCloud at the time of data processing. A first dataset was generated using a set of ten di- and tripeptides as the unknowns, the structure of the compounds was kept blinded to the analyst performing the work. MSn data acquired on these compounds was searched to identify the substructures and propose an identification. In all ten cases, the correct structure could be proposed. The utility of MSn data, combined with a suitably large and data rich MSn spectral library provides an additional tool to add to our repertoire for identifying the chemical structure of unknowns and can be a complementary approach to currently available capabilities.

Identification of substructures in small molecules utilizing an extensive MSn library

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.