Table 1 The total number of training particles-selection using fully automated good training particles-selection for apoferritin [27], KLH [26], and Ribosome [29] datasets
From: DeepCryoPicker: fully automated deep neural network for single protein particle picking in cryo-EM
Criteria | Apoferritin top-view | KLH top-view | KLH side-view | Ribosome irregular shape | β-Galactosidase complex particle shape |
|---|---|---|---|---|---|
Number of micrographs | 20 | 82 | 260 | 84 | |
Training micrograph | 10 | 65 | 208 | 67 | |
Testing micrograph | 10 | 17 | 52 | 15 | |
Size of micrograph | 1240 × 1200 | 2048 × 2048 | 4096 × 4096 | 4096 × 4096 | |
Resolution (Ã…) | 3.1 | 9.1 | 3.2 | 4.2 | |
Voxel size resolution (Ã…) | 0.82 | 1.24 | 1.34 | 1.77 | |
Total number of picked particles | 2145 | 1086 | 862 | 5493 | 8289 |
Particle’s patch size | 178 × 178 | 221 × 221 | 221 × 225 | 187 × 187 | 214 × 214 |
Number of good particles | 1750 | 887 | 689 | 1076 | 4781 |