Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Predicting Post-TOP Assay PFCA Burdens in Soil and Water Using Machine Learning: A Precursor-to-Product Modeling Approach.

Created on 13 Nov 2026

Authors

Euisung Oh, Sungjik Oh, Young Su Lee, Saerom Park, Qili Dai

Published in

Environmental research. Pages 126180. Nov 12, 2026. Epub Nov 12, 2026.

Abstract

The total oxidizable precursor (TOP) assay is widely used to reveal oxidizable per- and polyfluoroalkyl substance (PFAS) precursors that are not detected by conventional targeted analysis, but its labor- and cost-intensive nature limits large-scale monitoring. This study presents the first machine learning framework for predicting summed post-TOP perfluoroalkyl carboxylic acid (PFCA) concentrations from pre-oxidation PFCA profiles, enabling data-driven estimation of hidden precursor characteristics. Four tree-based ensemble models were trained using literature-derived datasets covering aqueous film-forming foam (AFFF) and non-AFFF contamination in water and soil. Six routinely measured PFCAs (C4-C9) were evaluated under unified, source-specific, and source- plus matrix-specific grouping strategies. Source-based grouping provided little improvement over the unified model (test R2 = 0.77-0.78), whereas incorporating environmental matrix substantially influenced prediction accuracy. The highest performance was achieved for non-AFFF water (R2 = 0.954) and AFFF-contaminated soil (R2 = 0.932), whereas lower predictive performance was observed for AFFF water and particular non-AFFF soil, indicating subgroup-dependent applicability. All tree-based ML models outperformed conventional linear baselines for the entire dataset (test R2 = 0.727-0.768 versus 0.396-0.410). Feature importance analysis suggested that PFHxA may be associated with fluorotelomer-derived precursors in water, while PFOA may be related to PFAS retention characteristics in soils. Performance was largely retained with three input PFCAs in the evaluated groups, suggesting a potential screening utility primarily for AFFF soil and non-AFFF water, while requiring caution for AFFF water and non-AFFF soil.

PMID:
43020411
Bibliographic data and abstract were imported from PubMed on 13 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 9
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement