Fig. 3: Training and picking by CASSPER and comparison with other tools. | Communications Biology

Fig. 3: Training and picking by CASSPER and comparison with other tools.

From: CASSPER is a semantic segmentation-based particle picking algorithm for single-particle cryo-electron microscopy

Fig. 3

a The validation F1 scores, which are used to select the best trained model in training epochs for prediction, for all three proteins. b Schematic representation of the uniform pipeline used to compare the performance of Gautomatch, CASSPER, and crYOLO. c Representative micrographs (scale bar, 50 nm) for (i) β-galactosidase, (ii) HCN1, and (iii) TcdA1 showing the particle picking performance of different tools. Highlighted areas indicate the noise picked by the respective tools. d Table showing the number of particles picked by Gautomatch, CASSPER, and crYOLO on the same micrograph for each protein as shown in c.

Back to article page