Idealizing Ion Channel Recordings by a Jump Segmentation Multiresolution Filter

Abstract

Based on a combination of jump segmentation and statistical multiresolution analysis for dependent data, a new approach called J-SMURF to idealize ion channel recordings has been developed. It is model-free in the sense that no a-priori assumptions about the channel’s characteristics have to be made; it thus complements existing methods which assume a model for the channel's dynamics, like hidden Markov models. The method accounts for the effect of an analog filter being applied before the data analysis, which results in colored noise, by adapting existing muliresolution statistics to this situation. J-SMURF’s ability to denoise the signal without missing events even when the signal-to-noise ratio is low is demonstrated on simulations as well as on ion current traces obtained from gramicidin A channels reconstituted into solvent-free planar membranes. When analyzing a newly synthesized acylated system of a fatty acid modified gramicidin channel, we are able to give statistical evidence for unknown gating characteristics such as subgating.

Unsupervised Idealization of Ion Channel Recordings by Minimum Description Length: Application to Human PIEZO1-Channels

“…We then tested the MDL method against a threshold-crossing algorithm, the Viterbi algorithm as implemented in QuB (Rabiner, 1989; Nicolai and Sachs, 2013), and a jump segmentation by multiresolution filter (J-SMURF) (Hotz et al, 2013). The test was performed under optimal detection conditions for the competing algorithms: For the threshold crossing algorithm, we used prior information of the unitary steps in the process to set the detection threshold to 0.5.…”

“…We first used “smuceR” from the R-package “stepR (Version 1.0)” to detect breaks on unfiltered data (Aspelmeier et al, 2016). In order to test jsmurf under similar conditions as in Hotz et al (2013) we low pass filtered the data using a 4 pole Bessel filter with cut-off = 0.05 to obtain SNR ~3. The filter characteristics were used as input parameter to the jsmurf function.…”

“…Hotz et al (2013) have used a jump-segmentation filter to analyze gramicidin A channels (J-SMURF) (Hotz et al, 2013). This method uses a statistical multiresolution analysis and detects abrupt changes in a recording, and does not assume a specific channel structure.…”

“…Our aim is to provide a fast and unbiased idealization of single channel time-series without requiring any user-dependent inputs. We assume that the current can be modeled stepwise as a sequence of segments of constant amplitude, separated by abrupt (instantaneous) transitions to some new current amplitude (Vandongen, 1996; Hotz et al, 2013). The algorithm identifies the location of the transitions between segments, and calculates the mean current prevailing in each segment.…”

Statistics and Related Topics in Single-Molecule Biophysics

“…The problem of change-point estimation has a long history in statistics dating back to the 1960s. More recent approaches, particularly relevant for single-channel data, include the use of BIC (Bayesian information criterion) penalty [57], quasi-likelihood method [58], L 1 penalty method [59], the multi-resolution method [60], and the marginal likelihood method [61]. Compared to the parametric inference methods based on continuous-time Markov chains, many of these change-point methods are flexible and can be made automatic.…”