Machine learning-assisted dual-mode immunochromatographic detection of pentachlorophenol via oriented antibody-functionalized AFIrTPs nanoprobes.

Abstract

Pentachlorophenol (PCP), a human carcinogen, is widely detected in human biosamples, despite being banned in aquaculture, continues to show high detection frequencies in aquatic products according to regulatory reports. Therefore, there is an urgent need for reliable on-site testing methods. Immunochromatographic assay (ICA) are promising, but conventional ICA suffers from antibody deactivation and unstable single-signal outputs in complex matrices. Herein, we reported a self-assembly method to fabricate tannic acid (TA)-mediated sea urchin-like Au/Fe/Ir trimetallic nanoparticles (AFIrTPs) that overcome these challenges. TA created a biofriendly interface that preserved Abs activity, while Fe3+ coordinated with the histidine-rich Fc domains to orient Abs, simplifying immobilization. Benefiting from the strong colorimetric response and efficient photothermal conversion of AFIrTPs, dual-mode ICA achieved detection limits of 3.47 pg/mL (colorimetric) and 2.19 pg/mL (photothermal). To facilitate PCP monitoring, a K-nearest neighbors algorithm enabled accurate classification and quantification of PCP, achieving a 99.3% classification accuracy. The assay shows satisfactory specificity and reliability, offering a promising platform for PCP on-site monitoring.