Alanine dipeptide
| Property | Value |
|---|---|
| Code | ACEMD |
| Forcefield | AMBER ff-99SB-ILDN |
| Integrator | Langevin |
| Integrator time step | 2 fs |
| Simulation time | 250 ns |
| Frame spacing | 1 ps |
| Temperature | 300 K |
| Volume | (2.3222 nm)^3 periodic box |
| Solvation | 651 TIP3P waters |
| Electrostatics | PME |
| PME real-space cutoff | 0.9 nm |
| PME grid spacing | 0.1 nm |
| PME updates | every two time steps |
| Constraints | all bonds between hydrogens and heavy atoms |
Contents
Raw data
PDB file and XTC files of three independent simulations (super-imposed configurations).
- alanine-dipeptide-nowater.pdb
- alanine-dipeptide-0-250ns-nowater.xtc
- alanine-dipeptide-1-250ns-nowater.xtc
- alanine-dipeptide-2-250ns-nowater.xtc
Featurized data
Each file contains three numpy.ndarray(shape=[250000, n_features], dtype=numpy.float32) objects (keys: arr_0, arr_1, arr_2) from three independent simulations.
- alanine-dipeptide-3x250ns-backbone-dihedrals.npz
- alanine-dipeptide-3x250ns-heavy-atom-distances.npz
- alanine-dipeptide-3x250ns-heavy-atom-positions.npz
Citations
- F. Nüske, et al: Markov State Models from short non-Equilibrium Simulations - Analysis and Correction of Estimation Bias, J. Chem. Phys. 146 (2017), 094104.
- C. Wehmeyer and F. Noé: Time-lagged autoencoders: Deep learning of slow collective variables for molecular kinetics, J. Chem. Phys. 148 (2018), 241703.