Wavelet Denoising of High-Bandwidth Nanopore and Ion-Channel Signals

Abstract

Recent work has pushed the noise-limited bandwidths of solid-state nanopore conductance recordings to more than 5 MHz and of ion channel conductance recordings to more than 500 kHz through the use of integrated complementary metal-oxide-semiconductor (CMOS) integrated circuits. Despite the spectral spread of the pulse-like signals that characterize these recordings when a sinusoidal basis is employed, Bessel filters are commonly used to denoise these signals to acceptable signal-to-noise ratios (SNRs) at the cost of losing many of the faster temporal features. Here, we report improvements to the SNR that can be achieved using wavelet denoising instead of Bessel filtering. When combined with state-of-the-art high-bandwidth CMOS recording instrumentation, we can reduce baseline noise levels by over a factor of four compared to a 2.5-MHz Bessel filter while retaining transient properties in the signal comparable to this filter bandwidth. Similarly, for ion channel recordings, we achieve a temporal response better than a 100-kHz Bessel filter with a noise level comparable to that achievable with a 25-kHz Bessel filter. Improvements in SNR can be used to achieve robust statistical analyses of these recordings, which may provide important insights into nanopore translocation dynamics and mechanisms of ion channel function.

Deep Learning‐Enhanced Nanopore Sensing of Single‐Nanoparticle Translocation Dynamics

“…The outputs displayed suppressed fast current fluctuations along with significant blunting of the I ion corrugations (Figure 3c, see also Figures S5 and S6, Supporting Information) ascribed to the fact that their time scale overlap with the frequency range of dominant noise components (Figure 3b). [ 39 ] These results manifest the aforementioned side effects of the conventional data processing to produce artifacts by deforming the signal shapes.…”

“…[ 38,39 ] However, previous studies have found it to be a non‐trivial task since the computation in frequency domains inevitably entails signal distortions thereby obscured the small yet important features occurring at variable time scales due to the stochastic and random nature of the translocation dynamics. [ 39 ]…”

“…[27,[34][35][36][37] Besides the material and device engineering for controlling the physical/chemical phenomena relevant to ionic current fluctuations, digital post processing has been utilized to mitigate the noise via band-pass filtering and wave transformations. [38,39] However, previous studies have found it to be a non-trivial task since the computation in frequency domains inevitably entails signal distortions thereby obscured the small yet important features occurring at variable time scales due to the stochastic and random nature of the translocation dynamics. [39] Here we report on a novel concept for post-denoising of ionic current in solid-state nanopores.…”

“…[38,39] However, previous studies have found it to be a non-trivial task since the computation in frequency domains inevitably entails signal distortions thereby obscured the small yet important features occurring at variable time scales due to the stochastic and random nature of the translocation dynamics. [39] Here we report on a novel concept for post-denoising of ionic current in solid-state nanopores. Our method is based on a deep learning algorithm formulated to extract noise floor from ionic current (I ion ) versus time (t) curves without clean data (Figure 1).…”

Detection of Apple Defects Based on the FCM-NPGA and a Multivariate Image Analysis

“…As can be seen from Table 1, the image denoising algorithm based on fractional difference has higher PSNR and SSIM compared to wavelet denoising [31] and mean filtering denoising [32]. It can be found by analyzing the data in the table that the denoising and edge-preserving capabilities based on the fractional difference algorithm are generally superior to those on the wavelet denoising and mean filter denoising.…”