EMReady: Post-Processed Cryo-EM Test Set Maps for Map Improvement

Aggregating cryo-electron microscopy (cryo-EM) test set maps post-processed using the EMReady deep learning framework. Developed by Florea Toma and Andreea V. based on 2023 research in Nature Communications, the collection provides maps enhanced through simultaneous local and non-local deep learning. It serves as a curated resource for evaluating methods and metrics in cryo-EM map refinement.

Use Cases

• Benchmarking map quality metrics against EMReady-enhanced density maps
• Evaluating structural biology visualization tools on deep learning-refined maps
• Comparing local and non-local deep learning improvements in protein structure maps

Strengths

• Derived from the EMReady method published in Nature Communications (2023)
• Features maps processed with simultaneous local and non-local deep learning architectures
• Curated specifically for benchmarking cryo-EM post-processing metrics

Limitations

• Unknown record count and file format specifications
• Lack of documentation regarding the original raw maps used as inputs
• No provided metadata columns for automated filtering by resolution or protein type

Provenance

Source

Florea Toma and Andreea V., based on He et al. (2023) Nature Communications

Collection Method

Post-processing of existing cryo-EM maps using the EMReady deep learning model

Freshness

Last updated March 2026.