Non-targeted analysis (NTA) methods use high-resolution mass spectrometry to better understand the identity of a wide variety of chemicals present in environmental samples (and other matrices). NTA methods are considered a qualitative analysis type, forgoing chemical standards (often because they are not available), and producing semi-quantitative concentration data using surrogate calibration techniques. Data collection for NTA experiments focuses on being as inclusive as possible without having a preconceived notion as to what chemicals would be of interest and allows retrospective queries of the data. Analysis of the resultant mass spectrometry information: 1) relies on cheminformatics to identify and rank chemicals, and 2) yields data that are used by cheminformaticians and decision makers to address environmental problems. In the first case, the CompTox Chemicals Dashboard is used to propose and select chemical structures supported by mass spectrometry data, either by single substances or batch searching using accurate mass/chemical formula. In the second case, NTA experiments produce lists of chemicals detected by a specific analysis type and suspected of being present in a particular environmental matrix. Those NTA data are then used to generate models to predict chromatography or ionization conditions likely to detect a particular molecule or the probability that similar compounds would be present in the same media. One source of such data is the EPA’s Non-Targeted Analysis Collaborative Trial (ENTACT), which involves the blinded and unblinded analysis of nearly 1,300 unique compounds combined into ten synthetic mixtures. The chemicals selected for ENTACT were originally procured for high throughput toxicity testing in the EPA’s ToxCast program. The wealth of information obtained from the analysis of the mixtures by at least sixteen laboratories is staggering. This information will be compiled into a publicly available database and will serve to improve our understanding of previously collected toxicity information, assist in optimizing analytical methods, and help predict the behavior of substances within the same chemical space We will report on progress in delivering the database for community access as well as on work to expand cheminformatics support for NTA such as providing tools and data to allow comparison of computationally generated fragmentation spectra for over 800,000 chemicals to experimentally acquired spectra.

Cheminformatics and non-targeted analysis: A two-way street

2019 Spring National Meeting

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