Fuzzy inference system as decision-maker to automate cryo-EM data acquisition on a transmission electron microscope

Abstract

Cryo-electron microscopy (Cryo-EM) data acquisition on modern transmission electron microscopes (TEM) is the first step during the single-particle analysis workflow. Importantly, the demand for large number of two dimensional images requires reliable and efficient automation of image data collection. We present a novel control scheme for automated cryo-EM data collection that monitors the quality of the data in real time and greatly improves the final efficiency of the acquisition. We propose the use of a fuzzy inference system (FIS) model to take decisions during the automated and sequential selection of hole areas in prefabricated EM grids. A new method based on adaptive neuro-fuzzy inference system (ANFIS) models was successfully trained to classify previously detected single particles from acquired images. In the methodology FIS and ANFIS are used to model expert behavior. The method is validated in real-time cryo-EM data acquisition for single-particle approach of bacterial ribosomes.