Data for "CryoDRGN: Reconstruction of heterogeneous cryo-EM structures using neural networks"

Trained models and reconstructed density maps for:

• EMPIAR-10028: "Cryo-EM structure of a Plasmodium falciparum 80S ribosome bound to the anti-protozoan drug emetine" from Wong et al. (2014)
• EMPIAR-10049: "Molecular Mechanism of V(D)J Recombination from Synaptic RAG1-RAG2 Complex Structures" from Ru et al. (2015)
• EMPIAR-10076: "Modular assembly of the large bacterial ribosome" from Davis et al. (2016)
• EMPIAR-10180: "Structure of a pre-catalytic spliceosome" from Plaschka et al. (2017)

Synthetic datasets with simulated heterogeneity and their ground truth density maps, poses, and labels:

• Uniform: 50k particle images (128x128, 6A/pix) uniformly sampled from 50 models along a 1-dimensional reaction coordinate
• Cooperative: 50k particle images (128x128, 6A/pix) sampled along the above reaction coordinate according to a 3-component Gaussian mixture model with overlapping components 
• Noncontiguous: 50k particle images (128x128, 6A/pix) sampled along the above reaction coordinate according to a 3-component Gaussian mixture model without overlapping components
• Ribosomes: 50k particle images (128x128, 3A/pix) containing a mixture of 30S, 50S, 70S ribosomes simulating compositional heterogeneity