Development and application of a retention time prediction model for more than 1500 new psychoactive substances - Forensics Webinar Week

Suspect screening based on Internet databases has proven to be a useful tool in overcoming the challenges of detecting rapidly emerging new psychoactive substances (NPS) in analytical toxicology. Unfortunately, these databases generally lack retention times specific to a user’s liquid chromatography (LC) setup and the subsequent application can result in an overwhelming number of false positives using mass spectrometry data alone. One such database is HighResNPS, an online crowd-sourced database which contains 3000+ entries corresponding to 1500+ unique NPS, including metabolites. Unlike most freely available databases, HighResNPS can be converted into formats supported by different LC-MS data analysis software.

Here we outline the development of a single retention time prediction model that can incorporate retention times recorded on multiple LC systems. This model can be used to predict the retention times for all unique entries on HighResNPS specific to those LC systems involved in the modelling process and as a result, “personalised” HighResNPS suspect screening databases can be created. We also demonstrate the application of the HighResNPS suspect screening database including predicted retention times using UNIFI.

Presenter: Daniel Pasin (Postdoctoral Researcher, Department of Forensic Medicine - Section of Forensic Chemistry - University of Copenhagen)

Daniel is currently a postdoctoral researcher in the Department of Forensic Medicine at the University of Copenhagen. His research interests involve the development of software-assisted strategies to enhance the detection capabilities of new psychoactive substances using high-resolution mass spectrometry. Currently, Daniel is investigating the use of machine learning techniques in combination with the programming language, Python, to develop models that can predict retention time and electrospray responses.

Presenter: Petur Weihe Dalsgaard (Senior Researcher, Department of Forensic Medicine - Section of Forensic Chemistry - University of Copenhagen)

Petur is currently a senior researcher in the Department of Forensic Medicine at the University of Copenhagen. He is responsible for the High Resolution Mass Spectrometry drug screening used in Forensic Toxicology at the Section of Forensic Chemistry. His research interests involve screening for new psychoactive substances, machine learning, and the development of internet databases such as HighResNPS.com