CRIM: Protein Structure Prediction Scores from Cryo-EM and IM-MS Integration

CRIM is an integrative scoring function for the Rosetta software that combines low-resolution cryo-electron microscopy density with ion mobility mass spectrometry-derived restraints to improve protein structure prediction. The dataset, authored by Jacob B. Howard and updated in May 2026, contains results from testing on an ideal set of 60 proteins and an experimental set of 54 proteins. It is a 12.9 KB XLSX file shared under a CC-BY-NC-4.0 license on figshare.

Use Cases

• Benchmarking integrative protein structure prediction methods based on the described CRIM scoring function.
• Comparing model selection performance against tools like AlphaFold3 based on RMSD and TM-score metrics.
• Training or validating machine learning models for structural biology using the provided experimental and ideal test set results.

Strengths

• Includes quantitative performance metrics (RMSD and TM-score) for 60 ideal and 54 experimental protein targets.
• Describes a specific, freely available tool (CRIM within Rosetta) with a clear methodological framework.
• Provides a direct comparison to a state-of-the-art method (AlphaFold3) for challenging targets.

Limitations

• Row count and specific column-level documentation are unknown; field semantics must be inferred after download.
• The dataset is very small (12.9 KB), indicating limited scope, likely containing summary results rather than raw structural data.

Provenance

Source

figshare, author Jacob B. Howard.

Collection Method

Results from computational modeling and scoring using the Rosetta software with the CRIM function.

Time Range

null

Freshness

Last updated 2026-05-14 06:05:34.

Geography

null