Proto is not affiliated with Institute for Protein Design. This toolkit is open source and builds on the implementation produced by this organization. Product names, logos, and trademarks are the property of their respective owners.

### RFdiffusion3 Structure Design (`rfdiffusion3-design`) Generates new protein structures and sequences subject to specified constraints. Each design task is described by an `RFdiffusion3DesignSpec` containing an optional input structure, a contig string, and per-residue selectors that fix atomic coordinates, constrain sequence positions, or designate hotspot residues. The diffusion sampler returns N candidate structures per specification, each accompanied by its designed amino-acid sequence and the sampled contig. #### API Reference

Source List of design specifications. Each spec represents an independent design task with its own constraints. Multiple specs will be processed in a single run. Input structure (`Structure` or PDB/CIF file path) for motif/binder design; written to a file before rfd3 (reads a path). Omit for de novo. Contig string specifying the design topology. Per-asymmetric-unit length (int or `"min-max"`); per-protomer under `symmetry`. Ligand selection by residue name from input structure. Comma-separated list of 3-letter codes (e.g., `"HAX,OAA"`). Unindexed motif components whose sequence position is unknown to the model. Useful for active site scaffolding where catalytic residues should be placed but their position in the final sequence is flexible. Components must not overlap with `contig`. Atoms to fix in 3D space during diffusion. Accepts InputSelection format: - `True`: Fix all atoms from input - `False`: Unfix all atoms - Contig string: Specify residues to fix (e.g., `"A1-10,B1-3"`) - Dict: Map chain/residue to atom selection using `"BKBN"` (backbone), `"TIP"` (tip atom), `"ALL"` (all atoms), or explicit atom names (e.g., `{"A1": "N,CA,C,O,CB", "A2-10": "BKBN"}`) Residues whose sequence can change during design. Accepts InputSelection format: - `True`: All input regions have fixed sequences (upstream default) - `False`: All atoms have unfixed/diffused sequences - Contig string: Components to unfix sequence for (e.g., `"A5-10,B1-3"`) Atom or residue hotspots for binder/PPI design (typically \<=4.5 Angstroms to any heavy atom in the designed structure). Symmetry for homo-oligomer design. A group-id string (e.g. `"C3"`) is wrapped as `{"id": "C3"}`; a full `SymmetryConfig` dict is passed through. Pair with `RFdiffusion3Config.sampler_kind="symmetry"`. RASA selector for buried residues. RASA selector for partially buried residues. RASA selector for solvent-exposed residues. Atom-wise H-bond donor flags, e.g. `{"A40": ["NE2"]}`. Atom-wise H-bond acceptor flags, e.g. `{"A45": ["OD1"]}`. Keep motif backbone fixed, redesign side-chains. Enable pLDDT-based denoising enhancement (upstream default `True`; `None` keeps it). Origin placement — `com` (center of mass) or `hotspots`. `[x, y, z]` origin override (Angstroms). Noise level (in Angstroms) for partial diffusion. Lower values preserve more of the input structure. Recommended values are 5.0-15.0 Angstroms. Useful for refinement or local redesign tasks. If `True`/`False`, produces output structures with fewer/more loops. `None` uses the model's native default. Raw JSON string for advanced users who need full RFdiffusion3 flexibility. If provided, `design_specs` will be ignored and this JSON will be passed directly to RFdiffusion3.

Source Independent batches per spec (total designs = `n_batches * diffusion_batch_size * num_specs`). Designs sampled in parallel per batch. Diffusion timesteps; more = slower, generally higher quality. Step size scale; higher = less diverse, more designable. Sampler kind; `'symmetry'` for homo-oligomer design (paired with `DesignSpec.symmetry`). Available options: `default`, `symmetry` Coordinate-frame centering — `all` (whole structure), `motif` (input motif), `diffuse` (diffused region only). Available options: `all`, `motif`, `diffuse` Enable CFG sampling. CFG scale factor (typical 1.0-3.0); no-op when CFG is off. Sampler stochasticity; lower = more designable, less diverse; `0.0` = deterministic ODE. Must be `> 0.5` when `sampler_kind="symmetry"`. Finer `inference_sampler` settings (noise schedule, motif noise); see that class for the fields. Diffusion noise scale (upstream default 1.003). Noise-schedule exponent (upstream default 7). Lower stochasticity bound, paired with `gamma_0` (upstream default 1.0). Translational noise scale for inference augmentation (upstream default 1.0). Allow motif realignment during sampling (upstream default False). Gaussian sigma jittering the motif offset (upstream default 0.0). Memory-efficient tokenization (slower); enable only if GPU RAM is tight. Save diffusion trajectory frames (debugging). Align trajectory frames across timesteps (only when `dump_trajectories=True`). Fail-fast input JSON validation. Checkpoint path or alias (`"rfd3"` = production preset). Local-execution input directory; `None` uses a tempdir. Local-execution output directory; `None` uses a tempdir. Verbosity level (0=quiet, 1=info, 2=debug, 3=raw subprocess stderr). `True` is coerced to `1` and `False` to `0`. `"cuda"` or `"cpu"`. Maximum execution time in seconds. `None` waits indefinitely. Random seed. When set, tools run reproducibly up to small GPU float noise (see `BaseToolOutput.approx_equal`), and the seed participates in cache keys. When None, cacheable seed-sensitive tools skip cache until seeded.

Source One bundle per input spec. Total design count is `len(design_specs) * n_batches * diffusion_batch_size`. Identifier of the input specification that produced these designs (e.g. `"spec-0"`). For `raw_json` callers this is the user-supplied key; otherwise it is positional (`f"spec-{i}"` from `RFdiffusion3Input.to_json_spec`). The designs generated for this spec, in the order RFdiffusion3 emitted them. List length equals `n_batches * diffusion_batch_size`.

#### Applications This tool is appropriate for any task in which the output is a novel protein backbone together with a designed amino-acid sequence subject to spatial constraints. Representative applications include protein-binder design against a target hotspot, motif scaffolding around catalytic or epitope residues, enzyme design around a small-molecule active site, symmetric homo-oligomer assembly, and partial-diffusion refinement of an existing backbone. RFdiffusion3 additionally supports nucleic-acid and small-molecule binder design within the same model and the same inference run, capabilities that the original RFdiffusion addressed only through separate model variants or task-specific extensions. #### Usage Tips * **One `RFdiffusion3DesignSpec` describes a single design task. Multiple specifications passed in one call are processed independently.** Each specification produces `n_batches * diffusion_batch_size` designs, and the resulting design sets are returned in input order. `length` is the only field that may be specified without an `input_structure`. The `contig`, `unindex`, and `select_*` fields are resolved against the input atomic coordinates and are rejected by validation when no input structure is provided. * **The `contig` string specifies the design topology.** Comma-separated segments select chain-indexed residues from the input (`A40-60`), insert designed regions of variable length (`70-80`), or introduce a chain break (`/0`). The full grammar is documented in the [RFdiffusion3 input specification](https://github.com/RosettaCommons/foundry/blob/production/models/rfd3/docs/input.md). * **`num_timesteps` and `step_scale` are the principal parameters controlling design quality and diversity.** Increasing `num_timesteps` (default `200`) extends the number of denoising steps and generally improves designability at a linear cost in runtime. `step_scale` (default `1.5`) scales the per-step denoising magnitude. Lower values yield more diverse but less designable outputs. Higher values produce structures closer to the training distribution. * **`sampler_kind="symmetry"` is required for homo-oligomer design and must be paired with a `symmetry` group on the specification.** Setting `symmetry="C3"` on the specification is converted internally to the `{"id": "C3"}` `SymmetryConfig` required by the RFdiffusion3 sampler. The symmetric sampler also imposes a `gamma_0 > 0.5` constraint, which is enforced during configuration validation. Under symmetry, `length` is per-protomer (`length="100"` + `symmetry="C3"` → 300-residue trimer). * **Classifier-free guidance is disabled by default and enabled by setting `use_classifier_free_guidance=True`.** When disabled, `cfg_scale`, `cfg_features`, and `cfg_t_max` have no effect. When enabled, increasing `cfg_scale` (typical range 1.0 to 3.0) more strongly weights the conditioning features (active-site hydrogen-bond donors and acceptors, per-atom relative accessible surface area) during sampling. * **`partial_t` enables partial diffusion, in which the sampler refines an input structure rather than starting from random noise.** The input is perturbed with Gaussian noise at the specified amplitude in angstroms and then denoised back, supporting local redesign and topology-preserving diversification of an existing backbone. The RFdiffusion3 `InputSpecification` reference lists 5.0 to 15.0 angstroms as a recommended range, but the [partial-diffusion documentation](https://github.com/RosettaCommons/foundry/blob/production/models/rfd3/docs/input.md#partial-diffusion) notes that `partial_t` is nonlinear and recommends beginning near 2 angstroms and increasing gradually. * **`is_non_loopy` is the only secondary-structure conditioning parameter available.** Setting it on `RFdiffusion3DesignSpec` to `True` biases the sampler toward fewer loops and substantially more helical content, with reduced sheet content. Setting it to `False` biases toward more loops and correspondingly fewer helices. Leaving it `None` (the default) applies no topology preference. The parameter is a single boolean flag and offers no per-residue or fractional control. * **Finer sampler tuning is grouped under the `sampler_tuning` field, a typed `RFdiffusion3SamplerTuning` whose settings map to the upstream `inference_sampler` block:** the noise schedule (`noise_scale`, `p`, `s_trans`), the stochasticity threshold `gamma_min` (paired with `gamma_0`), and the motif-noise settings `allow_realignment` / `s_jitter_origin` (effective only for motif-conditioned designs). Each defaults to `None` and is forwarded to the sampler only when set, otherwise inheriting the checkpoint's upstream default. Pass them as a dict, e.g. `RFdiffusion3Config(sampler_tuning={"noise_scale": 1.003})`; any other valid `inference_sampler` setting (e.g. `cfg_t_max`) may be included and is forwarded verbatim.