ThermoMPNN
ThermoMPNN Predicts protein stability changes with precision and efficiency for mutation analysis and design.
ThermoMPNN is a deep neural network designed to predict changes in protein stability resulting from point mutations, utilizing the protein's initial structure. By employing transfer learning, it leverages extensive datasets to enhance prediction accuracy. The model has demonstrated competitive performance on established benchmarks and is available as a tool for protein stability prediction and design.
Estimated resource cost: 0.001178
Categories: Protein Design
Tags: Developability Mutational Scanning Proteins Stability
Key Capabilities
- Predicts protein stability changes due to single or double point mutations.
- Features a Siamese neural network architecture for order-invariant double mutation predictions.
- Benchmarked against state-of-the-art models, demonstrating competitive or superior performance.
- Optimized for efficient predictions, with fast runtimes for single, additive, and epistatic modes.
Runtime Statistics
| Metric | Value |
|---|---|
| runtime_mean | 7 |
| runtime_median | 6 |
| runtime_std | 13 |
| runtime_90th_percentile | 13 |
| runtime_max | 205 |
Similar Tools
- BindFilter
- Efficient Evolution
- EpHod Optimal Enzyme pH Prediction
- BioBind
Ready to submit your job?
Review your configuration, then confirm the estimated credit cost before you run the job. Note that credit estimates are not guaranteed and runtime can vary depending on inputs and settings.
Estimated Credits: 0.001178
Invite-only, limited-time access. Please contact ztang@getantibody.com.
Invite-only, limited-time access. Please contact ztang@getantibody.com.