1. School of Biomedical Engineering(Suzhou)，Division of Life Sciences and Medicine, University of Science and Technology of China，Suzhou 215163，China
2. Suzhou Institute of Biomedical Engineering and Technology，Chinese Academy of Sciences，Engineering Research Center, Suzhou 215163，China
3. Chongqing Guoke Medical Innovation Technology Development Co., Ltd.，Molecular Diagnostic Center, Chongqing 400700，China
Abstract：Real-time fluorescence quantitative PCR is a commonly used detection method in molecular biology, mainly applied to detect DNA or RNA. However, the fluorescence data obtained by this method may feature crosstalk between fluorescence channels since there are overlapping fluorescence spectra and limitations of filter bandwidth. Such crosstalkcomplicates the PCR analysis and may ultimately affect the interpretation of detection results. Crosstalk between fluorescence channels can be reduced or eliminated by choosing appropriate filter combinations and using fluorescence crosstalk correction.Currently, the fluorescence crosstalk matrix is mostly estimated through aniterative algorithm, which is a complex method to obtain fluorescence crosstalk matrix from mixed multi-channel fluorescence data. A single dye experiment is carried out on the hardware platform to quickly calculate the fluorescence crosstalk matrix and reduce the computation. The principal component analysis (PCA) method is applied to estimate the distribution of dye fluorescence signals in each detection channel, and then the fluorescence crosstalk matrix is obtained. The crosstalk matrix shows that, for the built hardware platform, the Cy5 dye has a considerable crosstalk to the Cy5.5 channel with a crosstalk ratio of 8.76%; the Cy5.5 dye has a 6.2% crosstalk ratio to the Cy5 channel; the ROX dye has a 2.68%crosstalk ratio to the HEX channel; the crosstalk ratio of HEX dye to FAM channel is about 1.58%; the crosstalk ratio of FAM dye to HEX channelis relatively small, with only about 0.25%, and the other channels have no apparent crosstalk between each other, which is consistent with the fluorescence spectrum. The fluorescence crosstalk matrix is used to process the raw fluorescence data from the single dye experiment, which effectively removes the fluorescence data from the non-target channel and realizes the decoupling of the fluorescence channel data. The feasibility of the method is thus confirmed. Subsequently, a fluorescence separationexperiment is designed by randomly mixing various dyes of different concentrations to evaluate the quality of the crosstalk matrix's fluorescence correction. The experimental data are subject to fluorescence correction, and the linearity of the fluorescence for each dye is analyzed. The result demonstrates that the linear correlation of each fluorescent channel is high, and each linear correlation coefficient r of the five fluorescence channels exceeds 0.99, further validating the method's effectiveness.
[1] Brooke-Pearce F, Demertzi E. Journal of Infection Prevention, 2019, 20(6): 297.
[2] Zhu H, Zhang H, Ni S, et al. TrAC-Trends in Analytical Chemistry, 2020, 130: 115984.
[3] Huang E, Wang Y, Yang N, et al. Analytical and Bioanalytical Chemistry, 2021, 413(7): 1787.
[4] Oshiki M, Segawa T, Ishii S. Applied and Environmental Microbiology, 2018, 84(8): 1.
[5] Kimura Y, Ikeuchi M, Inoue Y, et al. Scientific Reports, 2018, 8(1): 17480.
[6] Lewis E K, Haaland W C, Nguyen F, et al. Proceedings of the National Academy of Sciences, 2005, 102(15): 5346.
[7] Geißler D, Stufler S, Löhmannsröben HG, et al. Journal of the American Chemical Society, 2013, 135(3): 1102.
[8] Liu Y, Wei M, Li Y, et al. Analytical Chemistry, 2017, 89(6): 3430.
[9] Huang W A, Yin Z B, Fuhrmann D R, et al. Electrophoresis, 1997, 18(1): 23.
[10] Huang W A, Fuhrmann D R, Politte D G, et al. IEEE Transactions on Biomedical Engineering, 1998, 45(4): 422.
[11] Li L, Speed T P. Electrophoresis, 1999, 20(7): 1433.
[12] Domnişoru C, Zhan X, Musavi M. Electrophoresis, 2000, 21(14): 2983.
[13] Gothot A, Grosdent J C, Paulus J M. Cytometry, 1996, 24(3): 214.
[14] ZANG Liu-qin, ZHANG Zhen-xi, MIAO Bao-gang, et al（臧留琴，张镇西，苗宝刚，等）. Acta Optica Sinica（光学学报）, 2014, 34(1): 0117002.
[15] Hasegawa T. Analytical Chemistry, 1999, 71(15): 3085.
[16] Hasegawa T, Nishijo J, Umemura J. Chemical Physics Letters, 2000, 317(6): 642.
