Jan Christensen on Quantitative Non-Targeted Screening

Non-targeted screening can flag thousands of features in an environmental sample, but without authentic standards it has been difficult to say how much of each compound is present. Predicted response factors aim to close that gap by estimating concentrations from models trained on compounds that have been measured with standards. At ISC 2026, Jan Christensen of the University of Copenhagen sat down with LCGC International to assess how far the approach has come and where it fits in an environmental screening workflow. Methods are improving a lot, Christensen says, but the challenge lies in the training data. Models need to be trained on a large number of compounds, and those compounds need to be similar to the ones being predicted. Some published studies rely on relatively few standards and then validate on the same set, or use leave-one-out approaches, which limits how much weight their reported performance can carry.

Christensen’s own experience, along with recent papers, points to better performance as training sets grow. Models trained on hundreds of compounds are now giving way to models trained on thousands. Christensen does not expect the approach to be used for regulatory purposes. Its value lies elsewhere: it gives analysts a way to rank the chemicals they detect by combining a concentration estimate with toxicity data. When a compound emerges as a potential high risk, it can then be followed up with a dedicated target analysis. For that kind of screening and prioritization, Christensen considers accuracy within a factor of two to three to be pretty good, and enough to turn a long list of detected features into a shorter list of chemicals worth a closer look.

Christensen answered the following question:

• How reliable is quantitative non-targeted screening using predicted response factors without authentic standards?

References

1. Zweigle, J.; Tisler, S.; Bevilacqua, M.; et al. Prioritization Strategies in Non-Target Screening of Environmental Samples by Chromatography with High-Resolution Mass Spectrometry. LCGC International 2026, 3 (2), 22–24. DOI: 10.56530/lcgc.int.hl3988l9
2. Christensen, J. H. Multidimensional Chromatography and LC-HRMS for Non-Target Screening of Environmental Contaminants: Strategies for Quantification, Prioritization, and Identification. Presented at ISC 2026, in Prague, Czech Republic. https://www.czech-in.org/cmPortalv15/Searchable/isc26/config/normal#!sessionschedule (accessed 2026-09-18).

Watch the full interview series here:

Jan Christensen on Multidimensional LC for Environmental Screening

Jan Christensen on Quantitative Non-Targeted Screening