You are here:
An open workflow to generate “MS Ready” structures and improve non-targeted mass spectrometry (ACS Fall 1 of 3)
Citation:
McEachran, A., K. Mansouri, Chris Grulke, AND A. Williams. An open workflow to generate “MS Ready” structures and improve non-targeted mass spectrometry (ACS Fall 1 of 3). Presented at 253rd American Chemical Society Fall National Meeting, San Francisco, California, August 20 - 24, 2017.
Impact/Purpose:
High-throughput non-targeted analyses (NTA) rely on chemical reference databases for tentative identification of observed chemical features. Many of these databases and online resources incorporate chemical structure data not in a form that is readily observed by mass spectrometry (e.g., structures that may contain salts, solvates, chemicals with stereochemistry included). Inclusion of MS-Ready structures within the Dashboard improved the coverage of candidate structures observed via HRMS, indicating the increased potential for identification of unknowns.
Description:
High-throughput non-targeted analyses (NTA) rely on chemical reference databases for tentative identification of observed chemical features. Many of these databases and online resources incorporate chemical structure data not in a form that is readily observed by mass spectrometry (e.g., structures that may contain salts, solvates, chemicals with stereochemistry included). Our center has previously reported on the production of “QSAR-ready structures”, forms of chemical structures processed for use in prediction models that are de-salted, de-solvated, and containing no stereochemistry information. While appropriate for property prediction and modeling purposes, QSAR-ready structures do not completely address structural issues of relevance to mass spectrometry. In this work we have adapted the workflow used to generate QSAR-ready structures to allow for the creation of “MS-Ready structures” that includes procedures to separate mixtures and multi-component substances within our database. The goal of this research is to provide an open workflow such that all users and architects of chemical reference databases have the capability to generate and query structures as observed in a mass spectrometer. Chemical structures were processed using KNIME Analytics Platform® and incorporated into the US EPA’s CompTox Chemistry Dashboard so that tentative identification is achievable by querying MS-Ready structures by monoisotopic mass and molecular formula. Inclusion of MS-Ready structures within the Dashboard improved the coverage of candidate structures observed via HRMS, indicating the increased potential for identification of unknowns. The availability of pre-processed MS-ready structure files as Open Data and the developing services for direct integration with other software tools positions the Dashboard as a valuable resource for the mass spectrometry community. This abstract does not necessarily represent the views or policies of the U.S. Environmental Protection Agency.