Related tools
EvoPro
Optimize protein binders using genetic algorithms combined with AlphaFold2 fitness evaluation and ProteinMPNN sequence design. EvoPro evolves protein sequences to maximize binding affinity and structural quality through iterative cycles of mutation, selection, and validation.
PepMLM
Design linear peptide binders for target proteins using a target sequence-conditioned masked language model. PepMLM generates peptide sequences optimized to bind specific protein targets based on ESM-2 protein language modeling.
BoltzGen
BoltzGen is a state-of-the-art AI model for designing protein and peptide binders against any biomolecular target. Using generative diffusion models, it creates novel binders (proteins, peptides, nanobodies) with nanomolar-level binding affinity.
mBER
Design VHH nanobody binders using AlphaFold-Multimer with structure templates and sequence conditioning. mBER (Manifold Binder Engineering and Refinement) generates novel VHH antibody sequences that bind to user-specified target proteins.
Humatch
Humatch is an antibody humanization tool that transforms non-human antibody sequences into humanized variants. Uses three lightweight CNNs to identify optimal human V-genes and generate paired heavy and light chain sequences with minimal edits while maintaining functionality.
IgDesign
Design antibody CDR sequences via inverse folding. Generates complementarity-determining region (CDR) sequences for antibodies targeting therapeutic antigens using deep learning. Optimizes CDR loops (HCDR1, HCDR2, HCDR3) based on antibody-antigen complex structures.
IgGM
IgGM is a generative foundation model for antibody and nanobody design against a target antigen. Supports CDR design, affinity maturation, inverse design, and framework design. Requires an antigen structure (PDB) and antibody sequences with "X" marking positions to design.
ODesign
All-atom generative AI for designing protein binders. Specify target binding sites and generate diverse binding proteins with fine-grained control over interaction parameters.
PocketFlow
PocketFlow is a structure-based molecular generative model that designs novel drug-like molecules within protein binding pockets. It uses autoregressive flow modeling with chemical knowledge to generate 100% chemically valid, highly drug-like compounds.
ProFam
ProFam-1 is a protein family language model for family-conditioned sequence generation. Provide a protein family FASTA/MSA and generate new sequences with model likelihood scores for downstream ranking and screening.
What is BindCraft?
BindCraft designs de novo protein binders against a target structure by combining AlphaFold2-guided backbone hallucination, ProteinMPNN sequence redesign, and PyRosetta interface scoring. It is built for cases where the goal is not to rank existing binders, but to generate entirely new mini-proteins that recognize a chosen surface patch.
The method is strongest on structured protein surfaces with clear geometric constraints. It is less reliable on highly flexible, disordered, or poorly resolved targets, where the hallucinated binder can optimize against an uncertain surface.
How to use BindCraft online
BindCraft on ProteinIQ accepts a target PDB structure plus chain selection, runs the upstream multi-stage binder-design pipeline in the cloud, and returns ranked binder-target complexes, full accepted-design metrics, and the remaining upstream CSV and artifact files for download in one job.
Inputs
| Input | Description |
|---|---|
Target Protein | One target structure in .pdb or .ent format. A structure can also be fetched from RCSB in PDB format. |
target_chains | Target chain ID or comma-separated chain list, for example A or A,B. Only the selected chains are used during design. |
hotspot_residues | Optional residue list such as A10,A15,A20. This biases the hallucination loss toward a chosen surface patch but does not absolutely force the final interface. |
Settings
| Setting | Description |
|---|---|
num_designs | Number of final accepted designs requested. Higher values increase runtime because more successful trajectories must pass filtering. |
length_min | Minimum binder length in residues. Shorter binders are faster to search but may not provide enough interface area for difficult targets. |
length_max | Maximum binder length in residues. Larger binders can cover broader interfaces but expand the search space. |
plddt_threshold | Displayed on the familiar 0-100 scale, default 80. Internally this is converted to the upstream normalized 0.8 threshold and written into Average_pLDDT, 1_pLDDT, and 2_pLDDT. |
ptm_threshold | Upstream normalized pTM filter threshold, default 0.55. Applied to Average_pTM, 1_pTM, and 2_pTM. |
i_ptm_threshold | Upstream normalized interface pTM threshold, default 0.5. Applied to Average_i_pTM, 1_i_pTM, and 2_i_pTM. |
design_algorithm | 4stage (default), 3stage, or 2stage. The four-stage preset is the upstream default and gives the most thorough optimization. |
use_multimer | Uses the AlphaFold2-multimer design preset, default true. This matches the upstream default_4stage_multimer.json configuration. |
enable_mpnn | Runs ProteinMPNN redesign after hallucination, default true. This usually improves sequence quality and folding behavior. |
Outputs
| Output | Description |
|---|---|
| Ranked complex PDBs | Primary outputs for the accepted binder-target complexes. The viewer uses these files directly. |
final_design_stats.csv | The accepted-design table produced by upstream BindCraft. ProteinIQ preserves this ranking and surfaces the full metric record in the Data tab. |
mpnn_design_stats.csv | Statistics for all ProteinMPNN redesign attempts, including accepted and rejected candidates. |
trajectory_stats.csv | Metrics for the hallucination trajectories before final sequence redesign. |
failure_csv.csv | Running counts of filter failures and rejection reasons across the job. |
| Remaining upstream artifacts | HTML trajectory animations, PNG plots, ZIP archives, FASTA files, JSON files, and text artifacts are returned when upstream leaves them in the output directory. |
How does BindCraft work?
AlphaFold2-guided hallucination
BindCraft uses AlphaFold2-Multimer as an optimization target rather than only as a predictor. The sequence is updated iteratively so that the predicted complex improves interface confidence, binder compactness, and other structural objectives.
Hotspot residues change the loss function, not the final scoring rule. A hotspot list can steer the search toward a preferred epitope, but the model can still drift toward a nearby surface if that geometry produces a more favorable predicted interface.
Multi-stage optimization
The 4stage protocol moves through progressively more discrete sequence representations. Early stages optimize softer sequence distributions, later stages enforce one-hot amino acid identities and mutation steps. The shorter 3stage and 2stage presets reduce compute cost by dropping some of that optimization path.
This matters most on difficult targets. Easy targets often produce acceptable binders under more than one preset, while hard interfaces benefit from the broader exploration provided by the four-stage search.
ProteinMPNN redesign and binder-alone validation
After a promising backbone is found, ProteinMPNN redesigns the binder sequence around that geometry. Upstream BindCraft then re-predicts the binder-target complex and also predicts the binder alone, which is why the accepted-design table includes both interface metrics and binder-only validation metrics such as Average_Binder_pLDDT, Average_Binder_pTM, Average_Binder_pAE, and Average_Binder_RMSD.
Filtering and relaxation
Upstream filtering is applied at multiple points. Early AlphaFold checks screen out weak trajectories, final acceptance combines the Average_* and model-indexed thresholds, and PyRosetta relaxation adds interface energetics and packing metrics such as Average_dG, Average_PackStat, Average_ShapeComplementarity, and Average_dSASA.
Understanding the results
The viewer shows a compact subset of fields for fast triage. The Data tab now contains the full accepted-design row from final_design_stats.csv, which is the best place to inspect why one design outranks another.
Metrics that usually matter first
| Metric | What it means |
|---|---|
Average_i_pTM | Predicted interface confidence. This is usually the first metric to inspect because it reflects whether the binder-target geometry looks self-consistent. |
Average_pLDDT | Overall local confidence across the predicted complex. Low values usually mean uncertain backbone geometry. |
Average_pTM | Global fold confidence for the complex. Useful as a stability check, but less interface-specific than Average_i_pTM. |
Average_dG | PyRosetta interface energy in REU. More negative values suggest a more favorable interface after relaxation. |
Average_ShapeComplementarity | How well the two surfaces fit together geometrically. Higher values generally indicate tighter packing. |
Average_PackStat | Packing quality of the interface. Higher is generally better. |
Average_dSASA | Buried solvent-accessible surface area on binding. Larger interfaces often bury more area, but large buried area alone does not guarantee a good design. |
Average_n_InterfaceResidues | Count of residues participating in the interface. Helps distinguish point contacts from broader binding patches. |
Average_Hotspot_RMSD | Distance from the designed binder pose to the intended hotspot geometry when hotspot targeting is used. Lower is better. |
Average_Binder_RMSD | How much the binder-alone prediction drifts from the complex-derived backbone. Large drift suggests the binder may depend on the target to stay folded. |
Practical interpretation
High-confidence designs usually combine strong Average_i_pTM, acceptable Average_pLDDT, and favorable interface energetics. A design with excellent Average_dG but weak Average_i_pTM often reflects a Rosetta-favored interface built on an uncertain AlphaFold geometry. A design with strong interface confidence but poor binder-alone metrics may bind only in the model because the isolated binder is unstable.
The upstream rank in final_design_stats.csv should be treated as the starting point, not the final answer. For experimental follow-up, the most useful shortlist often balances:
- Interface confidence from
Average_i_pTM - Structural confidence from
Average_pLDDT - Packing and energetics from
Average_ShapeComplementarity,Average_PackStat, andAverage_dG - Binder-alone stability from
Average_Binder_pLDDT,Average_Binder_pTM, andAverage_Binder_RMSD
When to use BindCraft vs alternatives
BindCraft is for de novo binder generation. It is not the right choice when a sequence already exists and only structure prediction, affinity ranking, or docking is needed.
| Tool | Best use case |
|---|---|
| BindCraft | Designing brand-new protein binders to a known target surface |
| ProteinMPNN | Sequence redesign for an existing backbone, without the full binder-hallucination pipeline |
| AlphaFold 2 | Structure prediction for an existing protein or complex sequence |
| DiffDock | Small-molecule docking, not protein-binder generation |
| AutoDock Vina | Physics-based small-molecule docking and pose ranking |