Fig. 5From: A deep convolutional neural network approach to single-particle recognition in cryo-electron microscopyTwo challenging examples of automated particle recognition. a A typical micrograph showing high-density top views of the inflammasome complex. Considerable ice contaminants and overlapping particles are present. b A typical micrograph of the side views of the inflammasome showing both a paucity of features and a low density of objects. The white square boxes indicate the positive particle images selected by DeepEM. The boxes with a triangle inside indicate that false-positive particle images were picked. The boxes with a star inside indicate the omitted particle images. c The precision-recall curves corresponding to the cases shown in (a) and (b)Back to article page