Simulation-guided optimization of one-step gold nanoparticle amplification for paper-based detection of anti-IFN-γ...

Adult-onset immunodeficiency (AOID) caused by anti-interferon-γ autoantibodies (anti-IFN-γ Abs) lacks rapid, sensitive diagnostic tools suitable for field use. A recent innovation-3D origami paper-based analytical devices (3D-osPADs) with gold nanoparticle amplification-offers a promising platform for single-step detection. This study uses computational modeling to validate and optimize the 3D-osPAD system, enhancing its diagnostic performance and scalability. We implemented a multi-step simulation pipeline encompassing calibration curve modeling, MES buffer-pH optimization, ROC and machine learning analysis under noise, Bayesian logistic regression for detection limits, vertical capillary flow modeling, and reagent stability estimation. All simulations were performed using open-source Python libraries in Google Colab. Simulations reproduced the 10 × sensitivity gain reported by Chien et al., reducing the detection limit from 0.01 to 0.001 μg/mL. MES-pH modeling identified an optimal amplification window at pH 5 and 150 mM MES. ROC analysis and ML classifiers (AUC >0.97) confirmed robust discrimination in noisy conditions. Bayesian modeling provided uncertainty-aware detection thresholds. Capillary flow modeling showed all six paper layers saturate in <1 s, while stability simulations highlighted the importance of cold-chain preservation for reagent integrity. Our modular in silico framework supports rapid optimization of 3D-osPAD diagnostics and provides actionable insights for low-cost, field-deployable biosensor development targeting anti-IFN-γ Abs and beyond. Our amplification occurs post-capture on immobilized Au labels and is therefore format-agnostic. We clarify how the same one-step growth can be integrated into standard lateral flow immunoassays (LFIA) without altering strip architecture or readout, thereby broadening practical applicability.