yzdai/md-trajectories

MD Trajectories of Phase-Coexisting Lipid Bilayers

Dataset Description

This dataset contains atomistic molecular dynamics (MD) trajectories of lipid bilayers exhibiting phase coexistence. The trajectories are used to validate phase recognition methods developed in our work[1] and cover a diverse set of membrane compositions, temperatures, and phase behaviors.

The dataset includes binary, ternary, and quaternary lipid mixtures, as well as bilayers containing membrane proteins. All simulations are performed at the microsecond time scale, providing sufficient sampling of phase separation and coexistence phenomena.

Systems Included

The simulation systems included in this dataset are summarized below.

Lipid Composition	# of Lipids	Protein	Temperature (K)	Simulation Time (μs)	Phase State
DPPC:DOPC = 3:2	1152	–	280	10	gel–Lα
DPPC:DOPC = 3:2	1152	–	290	10	gel–Lα
DPPC:DOPC:Cholesterol = 7:7:6	1152	–	280	8	Lo–Ld
DPPC:DOPC:Cholesterol = 7:7:6	1152	–	290	9	Lo–Ld
PSM:DOPC:POPC:Cholesterol = 45:36:9:38	512	–	300	20	Lo–Ld
DPPC:DOPC = 3:2	1082	AQP1	280	10	gel–Lα
DPPC:DOPC:Cholesterol = 7:7:6	1062	AQP1	280	10	Lo–Ld

Abbreviations

• DPPC: 1,2-dipalmitoyl-sn-glycero-3-phosphocholine
• DOPC: 1,2-dioleoyl-sn-glycero-3-phosphocholine
• POPC: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine
• PSM: palmitoylsphingomyelin

Simulation Methodology

All simulations were conducted using the GROMACS molecular dynamics package.

Binary and Ternary Bilayers (with/without AQP1)

• Force field (lipids): Slipids
• Force field (protein): AMBER14SB
• Water model: TIP3P
• Electrostatics: Particle Mesh Ewald (PME)
  • Cutoff: 1.0 nm
  • Fourier grid spacing: 0.12 nm
• Van der Waals interactions: Cutoff at 1.0 nm
• Temperature coupling: v-rescale thermostat (τ = 1 ps)
• Pressure coupling: Parrinello–Rahman barostat (semi-isotropic, τ = 3 ps)
• Bond constraints: LINCS (all bonds involving hydrogen)
• Ions: Added to neutralize protein-containing systems

Simulation protocols for binary and ternary bilayers without AQP1 follow those described in our previous work[2].

Quaternary Bilayers

• Force field: CHARMM36m
• Water model: TIP3P
• Electrostatics: PME
  • Cutoff: 1.2 nm
• Van der Waals interactions:
  • Force-switching between 0.8 and 1.2 nm
• Temperature and pressure coupling: Same as binary and ternary systems

Data Provenance

• Trajectories of binary and ternary lipid mixtures (DPPC:DOPC and DPPC:DOPC:Cholesterol) were obtained from our previous study[2].
• Trajectories of quaternary lipid mixtures and protein-containing bilayers were generated specifically for this work[1].

File Structure and Format

• Trajectory format: GROMACS compressed trajectory files (.xtc)
• Structure format: GROMACS structure files (.gro)

Citation

If you use this dataset in your research, please cite:

@dataset{yuzhuo_dai_2026,
  author       = { Yuzhuo Dai and Ruo-Xu Gu and SJTU },
  title        = { md-trajectories (Revision 9e56319) },
  year         = 2026,
  url          = { https://huggingface.co/datasets/yzdai/md-trajectories },
  doi          = { 10.57967/hf/7451 },
  publisher    = { Hugging Face }
}

When using the trajectories for phase identification or related analyses, we strongly encourage citing both the dataset DOI and the associated publication:

@article{dai2026recognition,
  title={Recognition of Coexisting Phases in Model Membranes via an Unsupervised Method},
  author={Dai, Yuzhuo and Zhao, Jianwei and Wang, Beibei and Liang, Qing and Gu, Ruo-Xu},
  journal={Journal of Chemical Information and Modeling},
  year={2026},
  publisher={ACS Publications}
}

References

[1] Dai Y, Zhao J, Wang B, Liang Q, Gu RX. Recognition of Coexisting Phases in Model Membranes via an Unsupervised Method. J Chem Inf Model. 2026;66(3):1840-1851. doi:10.1021/acs.jcim.5c02665

[2] Gu RX, Baoukina S, Tieleman DP. Phase Separation in Atomistic Simulations of Model Membranes. J Am Chem Soc. 2020;142(6):2844-2856. doi:10.1021/jacs.9b11057