Abstract
Surface enhanced Raman scattering (SERS) spectroscopy becomes increasingly used in biosensors for its capacity to detect and identify single molecules. In practice, a large number of SERS spectra are acquired and reliable ranking methods are thus essential for analysing all these data. Supervised classification strategies, which are the most effective methods, are usually applied but they require pre-determined models or classes. In this work, we propose to sort SERS spectra in unknown groups with an alternative strategy called Fourier polar representation. This non-fitting method based on simple Fourier sine and cosine transforms produces a fast and graphical representation for sorting SERS spectra with quantitative information. The reliability of this method was first investigated theoretically and numerically. Then, its performances were tested on two concrete biological examples: first with single amino-acid molecule (cysteine) and then with a mixture of three distinct odorous molecules. The benefits of this Fourier polar representation were highlighted and compared to the well-established statistical principal component analysis method.
Highlights
Biosensors are analytical devices measuring the concentration of specific analytes
As previously mentioned, when two or three molecules are passing through the Surface enhanced Raman scattering (SERS) active volume, the Fourier polar representation depicts a line with 3 or 4 poles (Fig. 3A)
We have introduced an alternative method for accurately and graphically sorting temporally fluctuating single molecule SERS spectra
Summary
Biosensors are analytical devices measuring the concentration of specific analytes. They are ubiquitous in many fields (e.g. for detecting pesticides[1,2], for analysing food quality[3,4] and environment[5] or for biomedical diagnostics[6,7,8]). When excited by an appropriate laser frequency, localized surface plasmon resonances sustained by the metallic nanoparticles provide a high enhancement of the electric field intensity that can reach several orders of magnitude (at least 106) In addition to their single molecule sensitivity, SERS sensors, by nature, provide a highly specific spectroscopic response at the individual molecules level[13,19]. The acquisition of a training set of samples These classification models are only valid for retrieving a known molecule with a given SERS based sensor and might not be adaptable to other platforms. An alternative sorting tool exists to group SERS spectra by similarities into unknown classes without the need of prerequisite models This multivariate analysis called principal component analysis (PCA) is used to reduce the dimensionality of measured SERS spectra into few principal components. This method is limited to disentangle relatively simple mixture because it is graphically bound to three principal components
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
