Suspect screening has proven to be a useful tool when trying to keep up with the constantly evolving chemical space. This has especially been the case for the emergence of new psychoactive substances (NPS), where substances can disappear from the drug market as quickly as they appeared, with each generation potentially yielding newer structural classes. This causes many logistical and financial problems for toxicology laboratories such as deciding which certified reference materials (CRMs) should be purchased for screening, confirmation and quantification.
While there are different intelligence-led processes that can be used to pragmatically inform a laboratory’s CRMS acquisition strategy (i.e. police seizure, wastewater analysis, drug checking and user survey data), they each have their own drawbacks. Suspect screening is a complementary tool that uses the acquired mass spectrometry data from a sample of interest to inform the decision-making process.
As a result, HighResNPS, the crowd-sourced mass spectral database for NPS, was developed by the Section of Forensic Chemistry at the University of Copenhagen and is now supported by the International Association of Forensic Toxicologists (TIAFT). As of March 2026, there are 8,526 entries corresponding to 2,470 compounds, including metabolites.
The database is converted into formats supported by most high-resolution mass spectrometry (HRMS) vendor software on a regular basis. Machine and deep learning algorithms have been applied to improve the utility of the database in suspect screening applications. This has included the development of a retention time prediction model which allows laboratories to download the database with predicted retention times that are specific to their LC conditions.