Benchmarks

Empirical performance metrics and throughput comparisons demonstrate MolSysMT’s scaling efficiency and comparative execution speed against industry-standard libraries (MDTraj, MDAnalysis, and SciPy).

Competitive Performance Timing Matrix

The table below presents comparative median execution durations across standard structural biology operations (evaluated on Trp-Cage miniprotein 1l2y and Chicken Villin HP35 datasets):

import pandas as pd
from IPython.display import HTML

data = [
    {"Operation Area": "Trajectory Load (DCD)", "MolSysMT Public": "176.95 ms", "MolSysMT Native Kernel": "N/A (Streaming)", "MDTraj": "29.65 ms", "MDAnalysis": "49.11 ms"},
    {"Operation Area": "Selection Simple (CA)", "MolSysMT Public": "5.83 ms", "MolSysMT Native Kernel": "N/A", "MDTraj": "3.95 ms", "MDAnalysis": "0.21 ms"},
    {"Operation Area": "Selection Complex", "MolSysMT Public": "10.54 ms", "MolSysMT Native Kernel": "N/A", "MDTraj": "58.33 ms", "MDAnalysis": "0.70 ms"},
    {"Operation Area": "Center of Geometry", "MolSysMT Public": "291.13 ms", "MolSysMT Native Kernel": "9.38 ms", "MDTraj": "1.79 ms", "MDAnalysis": "193.22 ms"},
    {"Operation Area": "RMSD Calculation", "MolSysMT Public": "306.49 ms", "MolSysMT Native Kernel": "8.34 ms", "MDTraj": "0.65 ms", "MDAnalysis": "177.92 ms"},
    {"Operation Area": "Pairwise Distances", "MolSysMT Public": "640,028 ms", "MolSysMT Native Kernel": "24,324 ms", "MDTraj": "4,649 ms (SciPy)", "MDAnalysis": "3,496 ms"}
]

df = pd.DataFrame(data)
html_table = df.to_html(classes="table", index=False)
html_table = html_table.replace('<th', '<th style="text-align: left;"').replace('<td', '<td style="text-align: left;"')
HTML(html_table)

Operation Area	MolSysMT Public	MolSysMT Native Kernel	MDTraj	MDAnalysis
Trajectory Load (DCD)	176.95 ms	N/A (Streaming)	29.65 ms	49.11 ms
Selection Simple (CA)	5.83 ms	N/A	3.95 ms	0.21 ms
Selection Complex	10.54 ms	N/A	58.33 ms	0.70 ms
Center of Geometry	291.13 ms	9.38 ms	1.79 ms	193.22 ms
RMSD Calculation	306.49 ms	8.34 ms	0.65 ms	177.92 ms
Pairwise Distances	640,028 ms	24,324 ms	4,649 ms (SciPy)	3,496 ms

Key Architectural Observations

• Complex Selections: MolSysMT’s selection parser resolves complex Boolean queries ((name CA or name CB) and resname ALA VAL LEU) in 10.5 ms, outperforming MDTraj’s bytecode compilation by 5.5x.

• Native Kernel Acceleration: Calling MolSysMT’s native compiled kernels directly achieves ~20x faster Center of Geometry and ~21x faster RMSD calculations compared to MDAnalysis wrappers.

• Public API Digestion Tax: The public API performs input validation, physical unit enforcement (pyunitwizard), and periodic boundary digestion. Controlled internal delegation may use skip_digestion=True only after the complete callee contract is established; MolSysMT has no value-passport protocol.