Detection of jumps in single-channel data containing subconductance levels

Abstract

Detection algorithms are widely used for the analysis of single-channel data because they remove the background noise from the measured current signal and reconstruct the noise-free time series. Standard detection algorithms assume channels switching only between zero and full conductance. Many types of channels, however, show subconductance levels. A new detection algorithm for data containing sublevels, the so-called sublevel Hinkley-detector (SHD), calculates several test values in parallel, one for each possible jump. The velocity of increase has a maximum for the correct jump. This feature is used to detect the jump and to diagnose the new level of current. Because patch-clamp data are always filtered by an antialiasing low-pass filter before sampling, the algorithm is supplemented by a special diagnosis phase accounting for the distortion of the originally rectangular jumps. Along with the reconstructed (noise-free) time series the SHD also gives a matrix of the transition counts between the levels. This matrix is a useful statistical tool for the decision whether the observed channel(s) have in fact a subconductance conformation or if there are simply several channels of different conductivity contained within the patch.