Data for "CryoDRGN-ET: Deep reconstructing generative networks for visualizing dynamic biomolecules inside cells"

Trained models weights, training parameters, sampled density maps, reconstructed density maps for featured classes, and plotting scripts are included for each of the following datasets and training runs: M. pneumoniae ribosome, initial training run with all 18,466 particles, 1 tilt per particle M. pneumoniae ribosome, training run with 16,655 filtered particles and 10 tilts per particle M. pneumoniae ribosome, training run with 16,655 filtered particles and 41 tilts per particle S. cerevisiae ribosome, initial training run with all 119,031 particles, 10 tilts per particle S. cerevisiae ribosome, training run with 93,281 filtered particles, 10 tilts per particle S. cerevisiae ribosome, training run with 30,657 particles in the non-rotated state, 10 tilts per particle S. cerevisiae ribosome, training run with 62,624 particles in the rotated state, 10 tilts per particle S. cerevisiae fatty acid synthase, initial training run with all 33,492 particles, 10 tilts per particle S. cerevisiae fatty acid synthase, training run with all 5,239 filtered particles, 10 tilts per particle