RFdiffusion2
Design functional enzymes from their reaction mechanisms using an atom-resolution generative model.
RFdiffusion2 is a deep generative method that enables the design of enzymes directly from sequence-agnostic, atom-level descriptions of their active sites (theozymes). It overcomes previous limitations by simultaneously inferring appropriate side-chain rotamers and sequence positions for catalytic residues, eliminating the need for computationally intensive enumeration steps.
Estimated resource cost: 0.000527
Categories: Protein Design
Tags: Enzymes Peptides Proteins
Key Capabilities
- Atom-level active site scaffolding
- Unindexed (sequence-agnostic) motif scaffolding
- Simultaneous rotamer and scaffold inference
- Conditioning on ligand functional groups
- Generation of full ligand conformers from partial structures
Runtime Statistics
| Metric | Value |
|---|---|
| runtime_mean | 1263 |
| runtime_median | 889 |
| runtime_std | 980 |
| runtime_90th_percentile | 2100 |
| runtime_max | 7221 |
Similar Tools
- AfCycDesign
- BindCraft
- BindFilter
- CatPred
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: 0.000527
Invite-only, limited-time access. Please contact ztang@getantibody.com.
Invite-only, limited-time access. Please contact ztang@getantibody.com.