Excitation-emission fluorescence matrix acquired from glutathione capped CdSeS/ZnS quantum dots in combination with chemometric tools for pattern-based sensing of neurotransmitters

Abstract

The presented work concerns pattern-based sensing with quantum dots for the identification and quantification of neurotransmitters by means of excitation-emission fluorescence spectroscopy (2D fluorescence). In the framework of this study, glutathione capped CdSeS/ZnS nanocrystals were used as non-specific nanoreceptors capable of differentiated interaction with neurotransmitters. The pattern-based sensing with QDs was realized by using excitation-emission fluorescence spectroscopy to provide analyte-specific multidimensional optical information. These characteristic fluorescent response patterns were processed by unfolded partial least squares–discriminant analysis, showing that satisfactory identification of all investigated neurotransmitters: dopamine, norepinephrine, epinephrine, serotonin, GABA, and acetylcholine, can be achieved through the proposed sensing strategy. The impact of the considered fluorescence signal (datum, i.e. zeroth-order data acquired per sample; spectrum, i.e. first-order data acquired per sample; excitation-emission matrix, i.e. second-order data acquired per sample) on the sensing capability of glutathione capped QDs was also verified. The best performance parameters such as accuracy, precision, sensitivity, and specificity were obtained using excitation-emission matrices (88.9–93.3%, 0.93–0.95, 0.89–0.93, and 0.99–1.00, respectively). Thus, it was revealed that excitation-emission fluorescence spectroscopy may improve the recognition of neurotransmitters while using only one type of nanoreceptor. Furthermore, is was demonstrated that the proposed excitation-emission fluorescence spectroscopy assisted QD assay coupled with unfolded partial least squares regression can be successfully utilized for quantitative determination of catecholamine neurotransmitters at the micromolar concentration range with R2 in the range 0.916–0.987. Consequently, the proposed sensing strategy has the potential to significantly simplify the sensing element and to expand the pool of bioanalytes so far detectable with the use of QDs. Graphical abstract

Feature Selection Assists BLSTM for the Ultrasensitive Detection of Bioflavonoids in Different Biological Matrices Based on the 3D Fluorescence Spectra of Gold Nanoclusters

“…Similarly, the RF-BLSTM model outperformed the BLSTM model by decreasing the value of RMSE by 70, 51, and 36% for the training, validation, and test set, respectively. Previous research indicates that the small values of MSE (and also RMSE) and the large value of R 2 are evidence that the model does not tend to overfit, and the model has excellent performance and accuracy. − Also, additional statistical analyses were shown in Tables S3 and S4, which indicate the superiority of the RF-BLSTM model over the BLSTM model.…”

Deep Learning-Based Sensor Array: 3D Fluorescence Spectra of Gold Nanoclusters for Qualitative and Quantitative Analysis of Vitamin B6 Derivatives

“…Also, the CNN model surpassed DNN by a decrease of 47, 7, and 33% of RMSE in the train, validation, and test set, respectively. The previous publications revealed that the lower value of MSE and RMSE and the higher value of R 2 led to better models’ performance and accuracy. , What is more, small values of RMSE (also MSE) indicated that the obtained models are not susceptible to overfitting . Thus, there may be a slight tendency for overfitting in the test set of DNN, where the values of MSE and R 2 are 81.05 and 93.28%, respectively.…”

“…62,63 What is more, small values of RMSE (also MSE) indicated that the obtained models are not susceptible to overfitting. 64 Thus, there may be a slight tendency for overfitting in the test set of DNN, where the values of MSE and R 2 are 81.05 and 93.28%, respectively. Altogether, these results indicate the superiority of CNN for qualitative and quantitative analysis of VB6Ds over DNN.…”

New Trends in Fluorescent Nanomaterials-Based Bio/Chemical Sensors for Neurohormones Detection─A Review

“…Głowacz et al 171 developed a strategy based on excitation–emission fluorescence spectroscopy using glutathione-capped CdSeS/ZnS quantum dots (QDs-GSH) for the detection of various neurotransmitters, including EP, NE, GABA, and more, because of the cross-affinity of the modified nanocrystals toward different chemical structures of neurotransmitters. The proposed assay allowed the quantification of catecholamine neurotransmitters (epinephrine, norepinephrine, and dopamine) at the micromolar concentration range.…”