MannnBeo/Dyn-PDB

Dynamic PDB protein Cα MD (processed)

A processed, protein-only view of the Dynamic PDB molecular-dynamics dataset (as deposited in MDRepo), made for training residue-level protein models. It is not the original dataset: the ligands, water and ions are removed and only per-residue quantities are kept. For the full simulations use the original sources below.

Source and licence

• Dynamic PDB: C. Liu et al., Dynamic PDB: A New Dataset and a SE(3) Model Extension by Integrating Dynamic Behaviors and Physical Properties in Protein Structures, AAAI 2025, https://arxiv.org/abs/2408.12413.
• Trajectories downloaded from MDRepo (https://mdrepo.org): the runs described as Dynamic PDB (not an MDRepo collection), one PDB chain each.
• This derived dataset is shared under CC BY 4.0 (MDRepo's terms); please cite the Dynamic PDB paper and MDRepo.

Simulations (from the MDRepo records): OpenMM, Amber ff14SB, TIP3P water, 150 mM NaCl; one 100-ns run per PDB chain at 300 K, frames every 1 ns.

What was changed

• Protein atoms only (mdtraj "protein" selection); the ligand, water and ions are dropped.
• Per residue and frame: Cα coordinates (Å), backbone φ and ψ (radians, NaN at chain ends), DSSP (8-state codes as int8); per frame: radius of gyration, end-to-end distance, replica (run_id) and time.
• Per residue and frame: the backbone N, C and O coordinates (Å), next to the Cα.
• Every frame of every replica is kept (no thinning).
• Train/validation split (3,145 / 165 proteins) by MMseqs2 sequence clusters at 30% identity: proteins of the same or similar proteins (e.g. one protein from several PDB entries) are always on the same side (1,412 / 115 clusters).

Contents

proteins (MDRepo runs)	3,310 of 3,365 Dynamic PDB runs in MDRepo
replicas per protein	1 (most); 0 proteins have fewer
frames	331,000
residue rows	59,286,000
chain length	9-598 residues, median 147

Runs left out:

reason	runs
longer than 600 (or shorter than 8) residues	55

Format

Plain NumPy files in dynpdb/ (download e.g. with huggingface_hub.snapshot_download(repo_id, repo_type="dataset")):

• ca.npy float32 [R, 3], phi.npy / psi.npy float32 [R], dssp.npy int8 [R]: every system's frames flattened residue-major, R = Σ T·L.
• bb.npy float32 [R, 3, 3] (Å): the backbone N, C, O (in that order) of each residue-frame, the rows of ca.npy (same frames and coordinates); NaN where a residue lacks the atom.
• rg.npy, e2e.npy, time_ps.npy float32 [F], run_id.npy int32 [F]: per frame, F = Σ T.
• index.json: systems, one entry per protein with name, seq, L, T, res_offset, frame_offset, split, cluster and meta (MDRepo id and slug, PDB id, UniProt ids, ligands, force field, frames per replica).
• SHA256SUMS: a checksum per file.

System s's frames: ca[s.res_offset : s.res_offset + s.T * s.L].reshape(s.T, s.L, 3).