With the help of machine learning, new techniques offer a promising approach to enhance single-molecule sensing with nanoparticle conjugates.
Researchers at the University of New South Wales and the Australian Center for Nanomedicine have developed a new technique that utilizes machine learning to analyze the plasmon resonance shifts of nanoparticle conjugates. A team led by J. Justin Gooding published its findings in the journal analytical chemistry (1).
Plasmonic nanoparticles are commonly used in single-molecule sensing, which are arranged in a dimer format. When the target molecule interacts with the hairpin DNA, the interparticle distance shifts, resulting in a localized surface plasmon resonance shift. This shift can be detected using spectroscopy, but it requires the measurement of thousands of nanoparticle dimers, is time consuming, and is not compatible with point-of-care devices.
To overcome this challenge, researchers used dark-field imaging of the dimer structure and used machine learning to analyze the plasmon resonance shift. By digitally separating the dimers from other nanoconjugate types, the team reduced false signals caused by clusters of non-specifically bound nanoparticles.
The team observed that variations in image intensity had a visible impact on the accuracy of color analysis and thus digital separation of nanoconjugate structures. To address this issue, the team compared different color spaces such as RGB, HSV, and LAB to train a classification algorithm. A LAB-based learning classifier showed the highest accuracy in digital separation of nanoparticles.
The team used a LAB-based learning classifier to monitor the plasmonic color shift of nanoparticle conjugates after interacting with synthetic RNA targets. The platform showed highly accurate yes or no responses with a true positive rate of 88% and a true negative rate of 100%. Sensor responses for the single-stranded RNA samples tested far exceeded control responses to target concentrations ranging from 10 aM to 1 pM.
In summary, this new technology offers a promising approach for enhancing single-molecule sensing with nanoparticle conjugates. The ability to achieve high throughput and statistical relevance in minutes makes this technology suitable for point-of-care devices that could revolutionize the field of diagnostic testing.
reference
(1) Bennett, D.; Chen, X.; Walker, GJ. Stelzer-Braid, S.; Rawlinson, WD; Hibbert, DB. Tilly, RD. Gooding, JJ Machine Learning Color Feature Analysis of a High Throughput Nanoparticle Conjugate Sensing Assay. anal. Chemistry. 2023, asap. DOI: 10.1021/acs.analchem.2c05292
