Machine Learning Predictions and Mechanistic Insights into Analyte Binding to DNA-Carbon Nanotube Sensors
ECS Meeting Abstracts · 07.07.2026 · MA2026-01 (9) · 789-789
DOI: 10.1149/ma2026-019789mtgabs ↗Kurz gefasst
Fachartikel in der Zeitschrift ECS Meeting Abstracts, erschienen am 07.07.2026. Verfasst von Lela Vukovic.
Abstract (Originalsprache)
Single-walled carbon nanotubes (SWNTs) wrapped with short single-stranded DNAs can function as near-infrared nanosensors of small biomolecules, yet the mechanisms of analyte binding and SWNT optical responses remain unclear. We combine molecular dynamics (MD) simulations and machine learning (ML) models to predict and explain sensing behavior in DNA-SWNT systems. Experimental data were obtained for 63 small molecules tested on (GT) 6 DNA-SWNT sensors, chosen to sample the chemical space around dopamine. Of these, 14 analytes spanning high, intermediate, and low responses were selected for MD simulations, while the full dataset was used to train ML models. MD simulations revealed distinct binding modes, including stacking, insertion, and DNA-backbone association. Analyte residence times and polar group-SWNT proximity correlated with experimental optical responses. Next, ML models trained on the experimentally obtained dataset identified chemical motifs linked to strong responses, and support vector models showed predictive capability, though limited by the dataset size. Blind predictions on 21 analytes demonstrated the feasibility of analyte-driven screening. Together, MD and ML approaches provide a complementary framework for mechanistic understanding and predictive design of DNA-SWNT nanosensors.
Zitieren
APA
Vukovic, L. (2026). Machine Learning Predictions and Mechanistic Insights into Analyte Binding to DNA-Carbon Nanotube Sensors. ECS Meeting Abstracts, MA2026-01(9), 789-789. https://doi.org/10.1149/ma2026-019789mtgabs
Vancouver
Vukovic L. Machine Learning Predictions and Mechanistic Insights into Analyte Binding to DNA-Carbon Nanotube Sensors. ECS Meeting Abstracts. 2026;MA2026-01(9):789-789. doi:10.1149/ma2026-019789mtgabs
BibTeX
@article{Vukovic2026,
author = {Vukovic, Lela},
title = {Machine Learning Predictions and Mechanistic Insights into Analyte Binding to DNA-Carbon Nanotube Sensors},
journal = {ECS Meeting Abstracts},
year = {2026},
volume = {MA2026-01},
number = {9},
pages = {789--789},
doi = {10.1149/ma2026-019789mtgabs}
}
EUVIMED