About EMInfo

EMInfo is a deep learning framework for detecting the secondary structures of proteins and the region of nucleic acids in a cryo-EM map.

Copyright © 2023 Hong Cao, Jiahua He, Tao Li, Sheng-You Huang and Huazhong University of Science and Technology. Released under GNU General Public License Version 3

Download EMInfo

EMInfo is freely available for academic and non-commercial users. [Download EMInfo v1.1]
If you have any questions regarding EMInfo, please don't hesitate to contact us at huangsy@hust.edu.cn

Reference
[1] Hong Cao, Jiahua He, Tao Li and Sheng-You Huang, Deciphering the structural information of proteins and nucleic acids in cryo-EM maps using deep learning 2023, submitted.

Explanations of output files:

example_c0.mrc: Prediction map of the helix regions of the protein, where 1 indicates helix and 0 indicates non-helix.
example_c1.mrc: Prediction map of the strand regions of the protein, where 1 indicates strand and 0 indicates non-strand.
example_c2.mrc: Prediction map of the coil regions of the protein, where 1 indicates coil and 0 indicates non-coil.
example_c3.mrc: Prediction map of the nucleic acid regions of the protein, where 1 indicates nucleic acid and 0 indicates non-nucleic acid.
example_all.mrc: Prediction map of the protein and nucleic acid regions, where 0, 1, 2, and 3 represent helix, strand, coil, and nucleic acid, respectively.
example_fold1.npy: Structure type prediction matrix for each voxel using the Fold1 model.
example_fold2.npy: Structure type prediction matrix for each voxel using the Fold2 model.
example_fold3.npy: Structure type prediction matrix for each voxel using the Fold3 model.
example_fold4.npy: Structure type prediction matrix for each voxel using the Fold4 model.
example_fold5.npy: Structure type prediction matrix for each voxel using the Fold5 model.