Examples » Convert to xtc format

In this example we read 2-nanosecond trajectory in .dat format and write as single .xtc file with various precision to compare different levels of lossy compression.

import os
from pyxmolpp2 import PdbFile, Trajectory, TrjtoolDatFile, XtcWriter

pdb_filename = os.path.join(os.environ["TEST_DATA_PATH"], "trjtool/GB1/run00001.pdb")
path_to_traj = os.path.join(os.environ["TEST_DATA_PATH"], "trjtool/GB1/")
frame = PdbFile(pdb_filename).frames()[0]


# Read trajectory and calculate net size in bytes
traj = Trajectory(frame)
dat_size = 0
for i in range(1, 3):
    dat_filename = f"{path_to_traj}/run{i:05d}.dat"
    traj.extend(TrjtoolDatFile(dat_filename))
    dat_size += os.path.getsize(dat_filename)


# Convert trajectory to .xtc and collect resulting sizes
sizes = {}
for precision in [100, 1000, 10000]:
    xtc_filename = f"trajectory_{precision}.xtc"
    xtc_writer = XtcWriter(xtc_filename, precision=precision)
    for frame in traj:
        xtc_writer.write(frame)
    del xtc_writer # closes file
    del frame # to re-enter trajectory
    sizes[precision] = os.path.getsize(xtc_filename)
    os.remove(xtc_filename) # clean-up

# Print results
print(f"                    .dat size: {dat_size:9d} bytes (100%)")
for precision, size in sizes.items():
    print(f".xtc size ({10/precision:5.3f} Å precision): "
          f"{size:9d} bytes ({size/dat_size*100:3.0f}%)")

                    .dat size:  21162264 bytes (100%)
.xtc size (0.100 Å precision):   4538048 bytes ( 21%)
.xtc size (0.010 Å precision):   6788108 bytes ( 32%)
.xtc size (0.001 Å precision):   8984660 bytes ( 42%)