Structure inpainting and simulation-ready setup for proteins, DNA, RNA, and complexes
Website | Atlas | Paper | PATCHR-Studio
Download PATCHR-Studio: Windows · macOS (Intel) · macOS (Apple Silicon) · Linux
 |
| PATCHR-Studio : end-to-end desktop workflow — load a structure, mark gaps, inpaint, and export |
 |
 |
| 1KX3 : nucleosome histone-tail inpainting | 6GIS : PCNA + 50 bp DNA extension |
 |
 |
| 8GZR : NS3 polymerase + RNA reconstruction | 4ZLO : Kinase inpainting with ligand |
Most experimental structures in the PDB have missing regions -- flexible loops, disordered terminals, unresolved sidechains. PATCHR fills them in using diffusion-based inpainting while keeping existing coordinates exactly as-is.
- Backend-agnostic -- supports Boltz-2 and Protenix
- Works with proteins, DNA, RNA, and multi-chain complexes
- 99.4% connectivity pass rate, from short loops to 600+ residue extensions
Benchmark — 940 PDB40 structures with artificially introduced gaps mirroring real PDB missing-region statistics. Cα and all-atom RMSD computed over inpainted residues only.
| Method / Configuration | Cα RMSD (Å) | All-atom RMSD (Å) |
|---|---|---|
| PATCHR (full, + LRD) | 1.781 | 2.542 |
| Boltz-2 + template conditioning | 4.647 | 5.510 |
| Boltz-2 + template conditioning + steering (threshold = 5.0 Å) | 3.675 | 4.342 |
| Boltz-2 + template conditioning + steering (threshold = 2.0 Å) | 3.397 | 4.081 |
| Boltz-2 + template conditioning + steering (threshold = 0.5 Å) | 3.219 | 3.889 |
| RFdiffusion2 (all-atom) | 9.188 | 10.199 |
| RFdiffusion (backbone-only) | 2.043 | — |
git clone https://github.com/DeepFoldProtein/patchr.git
cd patchr && pip install -e .Mac
conda create --name patchr python=3.12 llvmlite==0.44.0 numba==0.61.0 numpy==1.26.3
conda activate patchr
git clone https://github.com/DeepFoldProtein/patchr.git
cd patchr && pip install -e .
export KMP_DUPLICATE_LIB_OK=TRUEDocker
./scripts/docker-run.sh # Run with all GPUs
PATCHR_GPU=0 ./scripts/docker-run.sh # Select GPUModel weights are cached at ~/.boltz on the host (override with BOLTZ_CACHE).
For Slurm clusters with Apptainer:
sbatch scripts/slurm-run.sh1. Generate a template from a PDB structure:
patchr template 1TON all2. Run inpainting:
patchr predict examples/inpainting/1ton_AB.yaml --out_dir resultsThe first run downloads the model checkpoint automatically to ~/.boltz/.
Template options
patchr template 1CK4 all # All polymer chains
patchr template 4ZLO A,B --uniprot # With UniProt sequence
patchr template --input structure.cif A,B # From local CIF
patchr template 7EOQ all-copies # Including duplicate copies
patchr template 1BNA all -o my_templates/ # Custom output directory
patchr template 7EOQ A --include-solvent # Include solvent atoms
patchr template 1CK4 all --assembly best # Biological assembly
patchr template 1CK4 all --relative-paths # Use relative paths in YAML (default: absolute)Prediction options
# Single file
patchr predict examples/inpainting/4zlo_ABCD.yaml --out_dir results --seed 42
patchr predict examples/inpainting/1ck4_AB.yaml --out_dir results --diffusion_samples 5
patchr predict examples/inpainting/1bna_AB.yaml --out_dir results --backend protenix
patchr predict examples/inpainting/7eoq_ABCDEFGHIJKLMN.yaml --out_dir results --use_msa_server
# Bulk prediction — pass a directory of YAML files
patchr predict my_templates/ --out_dir results
patchr predict my_templates/ --out_dir results --backend protenix --seeds 42,101Go directly from structure completion to MD simulation input:
patchr predict input.yaml --out_dir results --sim-ready gromacs
patchr predict input.yaml --out_dir results --sim-ready amber --ff amber14sbStandalone command
patchr sim-ready prediction.cif --engine gromacs --ff charmm36m
patchr sim-ready prediction.cif --engine openmm --padding 1.2 --ion-conc 0.15PATCHR uses diffusion-based generation conditioned on your experimental structure as a rigid template:
| Technique | What it does | |
|---|---|---|
| 1 | Template Conditioning | Anchors known coordinates at every diffusion step |
| 2 | Synchronized Rigid Template Tracking | Keeps the template aligned with the evolving generation |
| 3 | Local Refinement Denoising | Cleans up bond geometry at template-generation junctions |
A desktop app with a visual interface for the full workflow -- no command line needed. Download from the links above or the releases page.
No GPU? Run the server on Google Colab for free and connect from PATCHR-Studio.
patchr serve --model boltz2 --device-id 0
patchr serve --model protenix --port 8080
patchr serve --model allPre-computed completed structures for ~35,000 monomeric proteins (growing to ~160,000 complexes). Explore the Atlas →
Beyond the headline RMSDs above, PATCHR also produces simulation-ready geometry:
| Metric | Value |
|---|---|
| Backbone RMSD (missing residues) | 1.78 Å |
| lDDT (missing atoms) | 98.6 |
| Connectivity pass rate | 99.4% |
Impact of Local Refinement Denoising (LRD)
| Metric | With LRD | Without LRD |
|---|---|---|
| Structures with no issues | 99.4% | 87.4% |
| Cα--Cα gaps (4.5--10 Å) | 0.21% | 4.57% |
| Peptide bond (C--N) issues | 0.85% | 15.43% |
| Broken chains (>10 Å) | 0.32% | 0.74% |
Accuracy by structural context
| Secondary structure | RMSD (Å) | Solvent accessibility | RMSD (Å) | |
|---|---|---|---|---|
| Helix | 0.30 | Buried | 0.39 | |
| Strand | 0.26 | Intermediate | 0.65 | |
| Loop | 0.85 | Surface | 1.01 |
Any model trained on the AlphaFold3 framework can be converted into an inpainting model through the PATCHR protocol. Currently implemented for Boltz-2 and Protenix only; extending to additional AF3-family backends is planned.
PATCHR builds upon Boltz-2 by Passaro, Corso, Wohlwend et al. and Protenix by ByteDance.
MIT -- free for academic and commercial use.
@article{bae2025patchr,
author = {Bae, Hanjin and Kim, Kunwoo and Yoo, Jejoong and Joo, Keehyoung},
title = {PATCHR-Studio: Template-conditioned diffusion-based molecular structure
inpainting for Protein, RNA, and DNA complexes},
year = {2025}
}