Unified All-Atom Molecule Generation with Neural Fields — MCPP Dataset
We curated a dataset of 186,685 MCP–protein complexes (mcpp_dataset.tar.gz) starting from 641 protein–MCP complexes from the RCSB PDB using a “mutate-then-relax” strategy:
We used this datase in FuncBind (see https://huggingface.co/papers/2511.15906).
Dataset Generation Pipeline
Mutation:
MCPs were randomly mutated at 1 to 8 sites using 213 distinct amino acids.Relaxation:
Mutated complexes were relaxed using FastRelax in Rosetta, which iteratively performs side-chain packing and all-atom minimization.Selection:
The best complexes were chosen based on lowest interface scores.
Dataset Statistics
- MCP lengths: 4–25 amino acids (average 10)
- 78% of MCPs contain one or more non-canonical amino acids
Dataset Splits
The dataset is split using a clustering-based approach. The test set covers 100 protein pockets:
| Split | File |
|---|---|
| Training set | train_data.pt |
| Validation set | val_data.pt |
| Test set | test_data.pt |
How to Use
- Download and extract:
tar -xvzf mcpp_dataset.tar.gz
- To generate MCP samples with Funcbind, :
cp train_data.pt val_data.pt test_data.pt mcpp_dataset/