MD Simulations
EXPLORATORYMolecular Dynamics (MD) simulations model how proteins move, fold, and interact with drug molecules over time. Each simulation runs on GPU hardware using OpenMM, tracking every atom at femtosecond resolution.
▶How does MD simulation setup work?
Each row represents a protein (or protein-drug complex) simulated under physiological conditions: 310 K (body temperature), explicit water solvent (TIP3P), 150 mM NaCl, periodic boundary conditions. Simulation types include SMN oligomerization, hnRNP A1–ISS-N1 binding, risdiplam mechanism, NCALD calcium dynamics, PLS3 actin bundling, and SMN–Gemin2 stability.
Key metrics. RMSD— < 2 Å plateau = stable fold; increasing = unfolding. Binding energy— < −7 kcal/mol = strong drug binding. Contact persistence — percent of time the drug maintains key interactions.
DiffDock predicts a static binding pose; MD tests whether that pose is dynamically stable. A drug that stays bound for 100 ns is a much stronger candidate than one that dissociates after 10 ns. GPU hours estimate compute cost on a single NVIDIA A100.
Completed Runs — actual GPU-fleet trajectories
29 runs · 140 ns total| Target | Ligand | PDB | ns | ns/day | Atoms | Final lig–prot Å | Force field | Completed | Data |
|---|---|---|---|---|---|---|---|---|---|
| KCNA2 | 4-aminopyridine | /data/research-data/SMA/_backup_archive/SMA/md_sims/4AP_Kv12/final_10ns.pdb | — | — | 64,801 | 1.95 | amber14SB | 2026-05-02 | — |
| HDAC2 | hydroxamate_rank01_518 | /data/research-data/SMA/_named_campaigns/hdac2_md_35136335/runs/rank01_518_hydroxamate/topology.pdb | — | — | 54,602 | 1.22 | amber14SB | 2026-05-02 | — |
| CORO1C | 4-aminopyridine_FEP | /data/research-data/SMA/_backup_archive/SMA/md_sims/4AP_FEP_CORO1C/final_10ns.pdb | — | — | 136,575 | 2.43 | amber14SB | 2026-05-02 | — |
| CORO1C | 4-aminopyridine_SMD | /data/research-data/SMA/_backup_archive/SMA/md_sims/4AP_SMD_CORO1C/final_10ns.pdb | — | — | 135,714 | 2.03 | amber14SB | 2026-05-02 | — |
| HDAC2 | hydroxamate_rank04_437 | /data/research-data/SMA/_named_campaigns/hdac2_md_35136335/runs/rank04_437_hydroxamate/topology.pdb | — | — | 54,688 | 1.06 | amber14SB | 2026-05-02 | — |
| HDAC2 | hydroxamate_rank15_200_bbb | /data/research-data/SMA/_named_campaigns/hdac2_md_35136335/runs/rank15_200_hydroxamate_bbb/topology.pdb | — | — | 54,698 | 1.16 | amber14SB | 2026-05-02 | — |
| PERP | design_H2b_9_s2 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/_smoke_H2b_9_s2/equilibrated.pdb | — | — | 149,782 | 19.97 | amber14SB | 2026-05-02 | — |
| PERP | design_H1c_25_s4 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/H1c_25_s4/equilibrated.pdb | — | — | 181,790 | 38.73 | amber14SB | 2026-05-02 | — |
| PERP | design_H2c_11_s1 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/H2c_11_s1/equilibrated.pdb | — | — | 218,812 | 26.20 | amber14SB | 2026-05-02 | — |
| PERP | design_H1c_25_s5 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/H1c_25_s5/equilibrated.pdb | — | — | 202,836 | 25.13 | amber14SB | 2026-05-02 | — |
| PERP | design_H1a_38_s7 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/H1a_38_s7/equilibrated.pdb | — | — | 147,483 | 29.79 | amber14SB | 2026-05-02 | — |
| PERP | design_H2b_9_s2 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/jak2_md_35124116_workspace/perp_md/traj/H2b_9_s2/equilibrated.pdb | — | — | 151,985 | 11.69 | amber14SB | 2026-05-02 | — |
| LIMK2 | sal_307_sal_pyrimidine_sel154 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/limk2_rock2_35138198_results/md_sims/LIMK2_LIMK2_307_sal_pyrimidine_sel154_holo_proper/solvated.pdb | — | — | 95,177 | 1.14 | amber14SB | 2026-05-02 | — |
| MDM2 | campaign_cmpd_cmpd1_mdm2v2 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/mdm2v2_cmpd1/equilibrated.pdb | — | — | 14,472 | 2.01 | amber14SB | 2026-05-02 | — |
| CFL1 | campaign_cmpd_cmpd2 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/cfl1_cmpd2/equilibrated.pdb | — | — | 151,878 | 23.89 | amber14SB | 2026-05-02 | — |
| MDM2 | campaign_cmpd_cmpd1_mdm2v1 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/mdm2v1_cmpd1/equilibrated.pdb | — | — | 15,121 | 1.32 | amber14SB | 2026-05-02 | — |
| MDM2 | campaign_cmpd_cmpd2_mdm2v1 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/mdm2v1_cmpd2/equilibrated.pdb | — | — | 15,209 | 1.20 | amber14SB | 2026-05-02 | — |
| CFL1 | campaign_cmpd_cmpd1 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/cfl1_cmpd1/equilibrated.pdb | — | — | 151,045 | 3.40 | amber14SB | 2026-05-02 | — |
| CFL1 | campaign_cmpd_cmpd3 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/cfl1_cmpd3/equilibrated.pdb | — | — | 150,994 | 25.52 | amber14SB | 2026-05-02 | — |
| MDM2 | campaign_cmpd_cmpd2_mdm2v2 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/cfl1_mdm2_35136325_workspace/md_run/traj/mdm2v2_cmpd2/equilibrated.pdb | — | — | 14,123 | 1.24 | amber14SB | 2026-05-02 | — |
| MUSK | diffdock_top1_id59 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/musk/md/MuSK_top1_id59/MuSK_top1_id59_equilibrated.pdb | — | — | 330,171 | 11.50 | amber14SB | 2026-05-02 | — |
| MUSK | diffdock_top3_id298 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/musk/md/MuSK_top3_id298/MuSK_top3_id298_equilibrated.pdb | — | — | 330,332 | 14.34 | amber14SB | 2026-05-02 | — |
| MUSK | diffdock_top2_id68 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/musk/md/MuSK_top2_id68/MuSK_top2_id68_solvated.pdb | — | — | 330,247 | 10.48 | amber14SB | 2026-05-02 | — |
| CDK5 | diffdock_top2_id22 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/cdk5/md/CDK5_top2_id22/CDK5_top2_id22_equilibrated.pdb | — | — | 28,496 | 1.44 | amber14SB | 2026-05-02 | — |
| CDK5 | diffdock_top1_id45 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/cdk5/md/CDK5_top1_id45/CDK5_top1_id45_final.pdb | — | — | 28,526 | 1.89 | amber14SB | 2026-05-02 | — |
| CDK5 | diffdock_top3_id369 | /data/research-data/SMA/_named_campaigns/md_backup_overnight_20260419/musk_cdk5_35136319_results/md/cdk5/md/CDK5_top3_id369/CDK5_top3_id369_equilibrated.pdb | — | — | 28,525 | 1.83 | amber14SB | 2026-05-02 | — |
| ROCK2 | — | 2F2U | 100 | 23.5 | 710,181 | — | amber14-all | 2026-04-11 | — |
| LIMK2 | BMS-5 | 4TPT | 20 | 167.7 | 95,608 | — | amber14-all + GAFF-2.11 (BCC charges) | 2026-04-11 | |
| LIMK2 | LIMKi3 | 4TPT | 20 | 167.7 | 95,608 | 7.59 | amber14-all + GAFF-2.11 (BCC charges) | 2026-04-11 |
KCNA2 + 4-aminopyridine: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
HDAC2 + hydroxamate_rank01_518: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CORO1C + 4-aminopyridine_FEP: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CORO1C + 4-aminopyridine_SMD: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
HDAC2 + hydroxamate_rank04_437: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
HDAC2 + hydroxamate_rank15_200_bbb: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H2b_9_s2: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H1c_25_s4: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H2c_11_s1: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H1c_25_s5: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H1a_38_s7: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
PERP + design_H2b_9_s2: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
LIMK2 + sal_307_sal_pyrimidine_sel154: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MDM2 + campaign_cmpd_cmpd1_mdm2v2: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CFL1 + campaign_cmpd_cmpd2: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MDM2 + campaign_cmpd_cmpd1_mdm2v1: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MDM2 + campaign_cmpd_cmpd2_mdm2v1: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CFL1 + campaign_cmpd_cmpd1: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CFL1 + campaign_cmpd_cmpd3: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MDM2 + campaign_cmpd_cmpd2_mdm2v2: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MUSK + diffdock_top1_id59: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MUSK + diffdock_top3_id298: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
MUSK + diffdock_top2_id68: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CDK5 + diffdock_top2_id22: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CDK5 + diffdock_top1_id45: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
CDK5 + diffdock_top3_id369: mdtraj backbone-RMSD analysis from Spark canonical (v2 topology-resolution)
ROCK2 + : Apo-ROCK2 100 ns baseline trajectory from PDB 2F2U. NO LIGAND BOUND despite the directory name mentioning CHEMBL38735 (COMPLETE marker confirms type=apo, compound=null). 4.26 days compute on RTX 3090, 710k atoms incl. solvent, stable at 300 K. Use as pocket-flexibility reference when comparing Fasudil / bbb5 / ROCK2 inhibitor MDs. NOT intended for Simon evidence package — this is a reference baseline, not a drug result. | Analysis 2026-04-12: RMSD artifact-inflated by N-terminal tails; core pocket Rg stable at 16.87A (CV 1.15%), DFG single-state throughout 100ns.
LIMK2 + BMS-5: Re-run with corrected GAFF-2.11 ligand parameterization to eliminate the +500 kcal/mol MMPBSA artifact from v1 distorted aromatic rings. Trajectory + checkpoint + final_20ns.pdb saved.
LIMK2 + LIMKi3: Re-run with corrected GAFF-2.11 ligand parameterization. 20 ns production. Final ligand-protein min distance 7.59 Å — ligand stays in pocket. Ready for MMPBSA.
Planned / Running Simulations
| Simulation | Target | Type | PDB / AF2 | Atoms | Time (ns) | GPU Hours | Force Field |
|---|---|---|---|---|---|---|---|
| SMN Protein Self-Oligomerization | SMN_PROTEIN | protein_protein | 4QQ6 | 45,000 | 100 | 8 | amber14-all.xml |
| hnRNP A1 binding to ISS-N1 RNA | SMN2 | protein_rna | 4YOE | 35,000 | 50 | 4 | amber14-all.xml |
| SMN2 pre-mRNA + Risdiplam binding | SMN2 | protein_ligand | - | 25,000 | 50 | 4 | amber14-all.xml |
| Neurocalcin-delta Calcium Binding Dynamics | NCALD | protein_ligand | AF-Q6P2D0-F1 | 30,000 | 200 | 16 | amber14-all.xml |
| Plastin-3 Actin Bundling Mechanism | PLS3 | protein_protein | 1AOA | 80,000 | 100 | 12 | amber14-all.xml |
| SMN-Gemin2 Complex Stability | SMN_PROTEIN | protein_protein | 2LEH | 50,000 | 100 | 8 | amber14-all.xml |
SMN Protein Self-Oligomerization
Simulate SMN Tudor domain self-association. SMN forms oligomers essential for Gems body formation and snRNP assembly. Understanding oligomerization dynamics reveals why reduced SMN (as in SMA) leads to functional collapse.
hnRNP A1 binding to ISS-N1 RNA
Simulate hnRNP A1 RRM domain binding to ISS-N1 (intron 7 +10 to +24). This is the interaction that nusinersen blocks. Understanding binding dynamics reveals small molecule disruption opportunities.
SMN2 pre-mRNA + Risdiplam binding
Simulate risdiplam (Evrysdi) binding to SMN2 pre-mRNA 5' splice site. Risdiplam stabilizes U1 snRNP binding, promoting exon 7 inclusion. Understanding the binding mode guides next-gen splicing modulator design.
Each simulation generates 3 Python scripts (setup, production, analysis). Run on GPU-equipped machines for best performance.