Deep-Channel uses deep neural networks to detect single-molecule events from patch-clamp data

Abstract

Single molecule research delivers a unique biological insight by capturing the movement of individual proteins in real time, unobscured by whole-cell ensemble averaging. The critical first step in single molecule analyses is event detection, so called "idealisation", where noisy raw data are turned into discrete records of protein movement. The most common type of single molecule research is electrophysiological patch-clamp recording of ion channel gating. To date, there have been practical limitations in the analyses pipelines for ion channel or other single molecule data; they are typically manually performed; laborious and require human supervision. In addition, this task can become infeasible with complex biological data containing many distinct native single ion channel proteins gating simultaneously. In this report we describe an "artificial intelligence" deep learning model for analyses of single molecule data, based on convolutional neural networks (CNN) and long short-term memory (LSTM) architecture. This network automatically idealises complex single molecule activity more accurately and faster than manual "threshold crossing" analyses. We believe this is the first use of deep learning to analyse single molecule datasets and such methods may revolutionise the unsupervised automatic detection of ion channel and other single-molecule transition events in the future.

Analysis of patchclamp recordings: model-free multiscale methods and software

“…Finally, an idealization can be obtained by the Viterbi algorithm (Viterbi 1967) or by Bayesian methods, in particular particle filtering, see (Fearnhead and Künsch 2018) and the references therein. Recently, a deep neural network approach has been proposed (Celik et al 2020), which skips the parameter estimation step and directly obtains an idealization. This approach can be seen as a hybrid method in between parametric and model-free approaches.…”

Unsupervised selection of optimal single-molecule time series idealization criterion

“…Unsupervised approaches for HMM model selection (e.g. infinite HMMs (Hines et al, 2015; Sgouralis & Pressé, 2017) and deep learning neural networks (Celik et al, 2020; Li et al, 2020; Xu et al, 2019) automate the process of model identification, but remain computationally expensive or require extensive training datasets prior to their use.…”

“…Compared to HMMs or change point analyses, DISC is orders of magnitude faster while maintaining state-of-the-art accuracy, precision and recall. Lately, many deep learning techniques reliant on neural networks have been developed for unsupervised SM analysis (Celik et al, 2020; Li et al, 2020). Unlike these approaches, DISC does not require extensive training datasets to guide its idealization which simplifies its application to multiple different experimental regimes.…”