关键词: Carboxylic acids Hierarchical cluster analysis Linear discriminant analysis Principal component analysis Pyruvic acid RGB images Visual sensor array

Mesh : Carboxylic Acids / analysis blood chemistry Cluster Analysis Colorimetry Copper / chemistry Discriminant Analysis Fluorescent Dyes / chemistry Humans Murexide / chemistry Nickel / chemistry Phenols / chemistry Principal Component Analysis Sulfoxides / chemistry Zinc / chemistry

来  源:   DOI:10.1007/s00604-019-3601-8

Abstract:
Carboxylic acids (CAs) have been reported as potential biomarkers of specific diseases or human body odors. A visual sensor array is described here that is based on indicator displacement assays (IDAs). The arrays were prepared by spotting solutions of the following metal complexes: Murexide-Ni(II), murexide-Cu(II), zincon-Zn(II) and xylenol orange-Cu(II), with the capability of discrimination of 15 carboxylic acids (CAs) and the quantitation of pyruvic acid (PA). Clear differences can be observed through distinctive difference maps obtained within 5 min by subtraction of red, green and blue (RGB) values of digital images after and before exposure to analytes. After an analysis of multidimensional data by pattern recognition algorithms including HCA, PCA and LDA, excellent classification specificity, and accuracy of >96% were obtained for all samples. The IDA array exhibited a linear range from 10 to 1500 μM with a theoretical detection limit of 3.5 μM towards PA. Recoveries of real samples varied from 84.8% to 114.3%. As-fabricated IDA sensor array showed an excellent selectivity among other organic interfering substances and a good batch to batch reproducibility, demonstrating its robustness. All these observations suggested that the IDA sensor array is one of the most promising paths for the discrimination of CAs. Graphical abstract Schematic diagram of indicator displacement assay (a), the procedure for acquisition of difference maps (b), and pattern recognitions for CAs (c). The method uses hierarchical cluster analysis (HCA), principal component analysis (PCA) and linear discriminant analysis (LDA).
摘要:
暂无翻译
公众号