niklastr/solvated_peptides

Neutral Peptide Ensembles

Molecular-dynamics ensembles combining matched neutral peptide samples from vacuum REMD, ManyPeptidesMD OBC1 implicit-solvent trajectories, and neutral microsolvated peptide frames. The vacuum and microsolvated ensembles are initialized from ManyPeptidesMD conformers.

Overview

Path	Model	Use
webdatasets/v1/joint_neutral/base_temperature/	Amber14 vacuum, Amber14/OBC1, and Amber14/TIP3P microsolvated clusters	300 K training
webdatasets/v1/joint_neutral/mixed_temperature/	The same environments with all vacuum REMD states	Temperature-conditioned training
trajectories/v1/joint_neutral/	Neutral test ensembles by environment	Evaluation

The vacuum and implicit_obc1 training components contain the same 4,398 formally neutral peptide sequences and exactly matching aggregate frame counts. The water component contains 2,729 neutral, no-counterion microsolvated peptide sequences.

Files follow:

trajectories/v1/joint_neutral/<environment>/<length>/<sequence>/...
webdatasets/v1/joint_neutral/<base_temperature|mixed_temperature>/<shard>.tar

The vacuum and water evaluation files use EqEnsemble NPZ files with positions, smiles, id, temperature, potential energy, and metadata. positions has shape (frames, atoms, 4) with rows [atomic_number, x_nm, y_nm, z_nm]. The implicit_obc1 evaluation files are ManyPeptidesMD NPZ trajectories containing positions, velocities, and sequence.

Each WebDataset sample has .bin, .smiles, .environment, and .json members. .bin is a float32 (atoms, 4) array with rows [atomic_number, x_nm, y_nm, z_nm]; .json contains source, environment, sequence, source frame, and temperature metadata.

Simulation

Starting structures for the vacuum and microsolvated ensembles came from the corresponding ManyPeptidesMD trajectories. Peptide sequences and train/test assignments follow ManyPeptidesMD.

Replica Exchange

Parameter	Value
Replicas	8, geometrically spaced from 300 to 500 K
Temperatures	300.00, 322.71, 347.14, 373.42, 401.69, 432.10, 464.81, 500.00 K
Integrator	Langevin, 2 fs, 0.3 ps^-1 collision rate
Warm-up / production	0.5 ns / 100 ns per replica
Save interval	5,000 steps = 10 ps
Exchange interval	5,000 steps = 10 ps

Vacuum and microsolvated REMD outputs were demultiplexed by thermodynamic state. The mixed-temperature WebDataset includes all eight states. Implicit OBC1 frames are uniformly subsampled from each 310 K ManyPeptidesMD trajectory to match the corresponding vacuum sequence's exported frame count. Water frames retain their source state temperatures.

Energy Function

• Vacuum: Amber14; NoCutoff; H-bond constraints.
• Implicit Solvent: Amber14 with OBC1; CutoffNonPeriodic at 2.0 nm; no constraints; 310 K.
• Explicit Solvent: Amber14 peptide in TIP3P. Distributed water frames contain the nearest 128 complete TIP3P waters and no counterions.

Frame Selection

Vacuum trajectories remove the first frame from every saved 201-frame simulation block, retaining 200 frames per block. Formal-charge neutrality is determined from the peptide SMILES. The microsolvated component additionally requires no counterions.

The base-temperature WebDataset contains 2,573,600 frames: 996,400 vacuum, 996,400 implicit OBC1, and 580,800 water. The mixed-temperature WebDataset contains 20,588,800 frames: 7,971,200 vacuum, 7,971,200 implicit OBC1, and 4,646,400 water.

Citation

Please cite this dataset and ManyPeptidesMD when using its peptide sequences, starting conformers, or OBC1 trajectories:

@misc{rindtorff_2026_matched_neutral_peptide_ensembles,
  author       = {Rindtorff, Niklas},
  title        = {Matched Neutral Peptide Ensembles},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/niklastr/solvated_peptides}},
  note         = {Dataset in progress; accessed 2026-07-23}
}

Source datasets: vacuum_peptides, microsolvated_peptides, and ManyPeptidesMD.

License

This derived dataset is released under CC BY 4.0. ManyPeptidesMD is distributed under the terms stated in its dataset repository. See the source repository for authoritative terms and provenance caveats.