基于小波新阈值函数的标记荧光蛋白信号去噪

针对微流控电泳仪采集的微弱标记荧光蛋白信号易受到干扰,影响对样品生物蛋白定性定量分析.提出可调新阈值函数小波去噪,解决了软、硬阈值函数去噪出现的常见问题.通过matlab仿真对比,寻找出去噪质量最佳db7小波基,结合可调新阈值函数处理重叠峰严重的标记荧光蛋白信号,去噪后波形最光滑、信噪比最高、获得最小均方根误差、更能突出真实信号.结果表明,新的阈值函数更适用于重叠峰严重的标记荧光蛋白信号,为后续定性定量分析样品中的生物蛋白含量提供保障.

De-Noising of Labeled Fluorescence Protein Signal Based on New Wavelet Threshold Function

The weakly labeled fluorescence protein signal collected by the microfluidic electrophoresis apparatus is susceptible to interference,which affects the qualitative and quantitative analysis of the biological protein in the sample.An adjustable new threshold function wavelet denoising was proposed to solve the common problems of soft and hard threshold function denoising.Through matlab simulation and comparison,the best db7 wavelet base for out-going noise quality was found,and the labeled new fluorescence protein signal was processed with the new adjustable threshold function.The denoised waveform was the smoothest,the signal-to-noise ratio was the highest,and the minimum root mean square error was obtained,which can better highlight the real signal.The results show that the new threshold function is more suitable for the labeling of fluorescence proteins with severe overlapping peaks,providing a guarantee for subsequent qualitative and quantitative analysis of biological protein content in samples.