version: "1.0.0" name: rfd4-proteina-generate description: > Run RFD4-Proteina (RFD4-Proteina) single-shot generation across every design task — the scientific anchor of the skill set. Reach for this skill for "design a protein", "generate a monomer", "run rfproteina inference", "scaffold a motif", "design an enzyme", "design a small molecule binder", "design a protein binder", "design a DNA binder", "fold / inverse-fold this backbone", "partial diffusion", "refine this structure", "make my designs more diverse", "deterministic (ODE) sampling", or "compute refolding metrics inline". It first runs the setup preflight, then routes by design case to the right benchmark example + overrides, and drives rfd4-generate. It hands off to rfd4-proteina-annotated-cif to build or edit an annotated CIF, to rfd4-proteina-search for reward-guided / beam search, and to rfd4-proteina-evaluate for deep metric analysis (the AF3 oracle). license: See the RFD4-Proteina repository (Baker Lab x NVIDIA collaboration)

RFD4-Proteina Generate

Single-shot generation for RFD4-Proteina. One entry point drives every design task — unconditional monomers/complexes, folding & inverse folding, motif scaffolding, enzyme design, small-molecule binders, protein binders, and DNA binders. The skill's job is to route the request to the right example input and the right overrides, then run one command. Deep input construction, reward-guided search, and oracle metrics live in sibling skills; this skill points at them.

Run everything from the repository root (the directory containing .project-root). All model arguments are Hydra key=value, never --key value.

Step 1 — Preflight

Never launch a long run against a broken environment. Confirm venv, GPU, and checkpoints first:

bash skills/rfd4-proteina-setup/scripts/preflight.sh

Fix every FAIL (see rfd4-proteina-setup) before generating. scripts/download_assets.sh fetches the release pair composed by configs/inference.yaml, so the default run needs no checkpoint override.

Step 2 — Route the task

Pick the design case, then open its reference for the exact command, example input, task-specific overrides, and gotchas:

Need an input the request doesn't have? Annotated CIF is the recommended, lossless format. When the task needs coordinate/sequence conditioning, hotspots, fixed ligands, H-bonds, RASA, or unindexed motifs, hand off to rfd4-proteina-annotated-cif to build and validate the CIF, then come back and pass it via inputs=<file>.cif.

Step 3 — Run

General form:

rfd4-generate inputs=<file> num_replicates=<N> out_dir=<dir> [overrides...]

Use the Python environment activated during setup. The installed command runs in that environment; see the installation guide for environment and scheduler settings.

The four parameters you almost always set (defaults are only right for a smoke test):

Useful extras: subset_to_examples=monomer-short (run only named keys), seed=<int> (reproducibility), skip_existing=true (resume a run), save_chain_of_interest=true (required for later foldseek clustering).

# 80 designs of each monomer example, 8 at a time
rfd4-generate \
    inputs=data/benchmarks/monomer \
    num_replicates=80 batch_size=8 out_dir=./runs/monomers

Scheduler intent

For any batch run, read the selected scripts/.user_info_<cluster> and the launcher before constructing the command. Preserve the user's partition, account, QoS/priority options, GPU request, runtime, memory, and CPU settings. Use a resource override only when the user explicitly requested it. State the resolved scheduler values before submission; use dry-run or non-submitting tests when validating the workflow.

Step 4 — Pick your overrides

Full cheat sheet: references/overrides.md. The knobs that matter most:

• Temperature / diversity. Use

generation.model.bb_ca.simulation_step_params.tsr_k (default 16) and generation.model.local_latents.simulation_step_params.tsr_k (default 4). Higher k lowers temperature; lower k raises it. tsr_sigma provides secondary control over when sharpening takes effect (defaults: backbone 3, latent 12).

• Partial diffusion (refine an existing structure). Simple:

generation.args.partial_trajectory_fraction_steps=<f> (fraction of nsteps). Advanced: generation.args.partial_t_backbone=<t> / generation.args.partial_t_latent=<t> (floats in (0,1); near 1 stays close to the input, near 0 approaches full generation), optionally generation.args.couple_partial_t=true. partial_t is a corruption level, not Angstroms.

• Deterministic (ODE) sampling (no mid-trajectory noise):

generation.model.bb_ca.gt.mode=ode generation.model.local_latents.gt.mode=ode.

• Steps / recycling. Use the task schedule table.

All presets use 60 steps; select the task's medium unless another budget was requested. thinking_* aliases are front-loaded, not task-aware selectors.

Step 5 — Inline metrics (optional)

Pass metrics_tags='[...]' to compute metrics right after each design; results land in per_example_metrics.csv alongside the CIFs. Common tags include design, unconditional_refolding, ppi_refolding, dna_refolding, motif_scaffold_refolding, sm_binder_refolding, ame_refolding, disco_refolding, hotspot_conditioning, 3d_coordinate_conditioning, rasa_conditioning, and interchain_interactions. Check docs/how-to/evaluate-with-metrics.md and the composed metric configs before treating any list as exhaustive.

[!NOTE]

Every *_refolding tag runs the AF3 oracle, which needs an apptainer/docker

container (see rfd4-proteina-setup for the container, rfd4-proteina-evaluate

for what the metrics mean and how to read them). Plain generation needs no

container. Deep metric detail belongs to rfd4-proteina-evaluate — this skill

only turns the tags on.

rfd4-generate \
    inputs=data/benchmarks/monomer num_replicates=50 batch_size=8 \
    out_dir=./runs/monomers_scored metrics_tags='[unconditional_refolding]'

Reward-guided / beam search

For "generate many, keep the best under an AF3 reward", do not hand-roll it here — use rfd4-proteina-search, which drives configs/inference_experiment/inference_beam_example.yaml and the search config knobs (beam width, branching, checkpoints, metric weights).

Reusable configs

For a repeatable experiment instead of one-off CLI overrides, copy configs/inference_experiment/TEMPLATE.yaml — it documents the three ways to run (in-tree experiment config, CLI overrides, config file anywhere on the filesystem). The default group is inference_on_user_inputs.

Outputs

Each design is an annotated CIF with per-atom provenance columns (SA_SRC, SA_CRDSRC, SA_IDXGRPSRC, SA_UII) tying output atoms back to the input spec. With metrics on, a per_example_metrics.csv is written alongside. Full column table and how to read annotations: references/outputs.md.

Troubleshooting

For deeper detail: overrides · outputs · per-task guides under references/tasks/.