Automated Stoichiometry Analysis of Single-Molecule Fluorescence Imaging Traces via Deep Learning

Abstract

The stoichiometry of protein complexes is precisely regulated in cells and is fundamental to protein function. Singe-molecule fluorescence imaging based photobleaching event counting is a new approach for protein stoichiometry determination under physiological conditions. Due to the interference of the high noise level and photoblinking events, accurately extracting real bleaching steps from single-molecule fluorescence traces is still a challenging task. Here, we develop a novel method of using convolutional and long-short-term memory deep learning neural network (CLDNN) for photobleaching event counting. We design the convolutional layers to accurately extract features of steplike photobleaching drops and long-short-term memory (LSTM) recurrent layers to distinguish between photobleaching and photoblinking events. Compared with traditional algorithms, CLDNN shows higher accuracy with at least 2 orders of magnitude improvement of efficiency, and it does not require user-specified parameters. We have verified our CLDNN method using experimental data from imaging of single dye-labeled molecules in vitro and epidermal growth factor receptors (EGFR) on cells. Our CLDNN method is expected to provide a new strategy to stoichiometry study and time series analysis in chemistry.

Flapping Peryleneimide as a Fluorogenic Dye with High Photostability and Strong Visible‐Light Absorption

“…It simultaneously displays highly sensitive FL lifetime response, strong absorption in the visible region, and high photostability, which will open up new possibilities in long-time and super-resolution microscopy observation for mapping local environments and/ or molecular communications. [17,18] Since synthetic approaches to water-soluble perylene dyes have been already established, [15] the flapping peryleneimides also have a potential for biological applications as sensitive fluorescent probes in a low-viscosity region. [26]…”

Photobleaching step analysis for robust determination of protein complex stoichiometries

“…We believe that the developed method of testing algorithms with semi-synthetic data will be highly useful, not only for benchmarking other PBSA algorithms, but also to generate training data for machine learning based approaches. 20 Using ATTO 647N-labelled structures with a known stoichiometry in vitro, we showed that quickPBSA yields highly accurate (<10% deviation across all samples) estimates of All data was background corrected by subtracting a constant offset from acquired image series to account for excitation light bleedthrough and sample autofluorescence. Offsets were manually determined for each sample and were well reproducible within one condition, but varied between conditions.…”

“…Recently molecular counting via photobleaching has attracted renewed attention due to the development of novel analysis modalities based on Bayesian statistics 18,19 and machine learning. 20 But these novel methods are demanding in terms of data quality, which in turn leads to new requirements regarding fluorophore properties. In the trade-off between signal-tonoise ratio (SNR) of the individual bleaching steps and the rate at which photobleaching occurs, bright and stable fluorophores are advantageous.…”

Logic‐Gate‐Actuated DNA‐Controlled Receptor Assembly for the Programmable Modulation of Cellular Signal Transduction

“…[2] Ty pically,t he input sensors have attracted considerable interest as processing units to detect and integrate multiple environmental signals. [3] Multiple-input logic operations are necessary for cells to ensure the accuracy of initiating ad esired response.F or example,t he beta cells regulate the secretion of insulin in response to different stimuli, such as glucose,g lucagon and hormones,a nd thereby maintain glucose homeostasis. [4] Thus,t here is al ong-standing interest in reprogramming cells to sense multiple-inputs and initiate specific cellular signal transduction as am imic of natural complex information processing.…”