Automatic identification of crossovers in cryo‐EM images of murine amyloid protein A fibrils with machine learning

Abstract

Detecting crossovers in cryo-electron microscopy images of protein fibrils is an important step towards determining the morphological composition of a sample. Currently, the crossover locations are picked by hand, which introduces errors and is a time-consuming procedure. With the rise of deep learning in computer vision tasks, the automation of such problems has become more and more applicable. However, because of insufficient quality of raw data and missing labels, neural networks alone cannot be applied successfully to target the given problem. Thus, we propose an approach combining conventional computer vision techniques and deep learning to automatically detect fibril crossovers in two-dimensional cryo-electron microscopy image data and apply it to murine amyloid protein A fibrils, where we first use direct image processing methods to simplify the image data such that a convolutional neural network can be applied to the remaining segmentation problem.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Structural studies of elastic fibre and microfibrillar proteins

“…Technologies increasing resolution and contrast of samples make cryoEM suitable for studying smaller targets, such as proteases and regulators, controlling ECM formation and reorganisation, whilst novel methods to conduct time-based experiments allow the study of dynamic complexes. Improved image classification techniques are increasingly capable of sub-classifying individual conformational states in heterogeneous samples, which makes it possible to study complexes and processes with multiple different functional states [146] , whilst other algorithms are able to reliably detect non-globular particles which is vital for the study of fibrillar ECM components [147] , [148] . As a result, cryoEM still holds large potential for further advances in structure-based research, beyond traditional structural biology and makes cryoEM one of the driving technologies to improve our understanding of the structure and function of ECM proteins, complexes and processes.…”

Machine learning in electron microscopy for advanced nanocharacterization: current developments, available tools and future outlook

“…However, it is important to keep it in mind when trying to apply ML to materials science (or to any other field), as similar routines might have already been developed, for instance, within the mentioned cryo-TEM community. [177][178][179][180][181][182][183][184][185][186][187][188][189][190][191][192] The cross-fertilization with not only other microscopy techniques, but other scientific and technical disciplines is of capital importance and is deeply discussed in the fourth section of this review.…”

Challenges in sample preparation and structure determination of amyloids by cryo-EM

“…2). An alternative and less laborious approach to manual selection, is the automated detection of cross-overs in images of amyloid fibrils, which can be achieved by applying conventional computer vision techniques combined with machine learning approaches (159). This method enables the statistical analysis of the sample and thus, gives insights into its morphological composition.…”