Trajectory#

Define the ways to represent a trajectory.

class nearl.io.traj.Trajectory(traj_src=None, pdb_src=None, **kwarg)[source]#

Bases: Trajectory

This class represents the base-class for trajectory handling in the Nearl package. This class inherits from pytraj.Trajectory.

Notes

Three types of trajectory initialization are supported:

File-based trajectory initialization (traj_src and pdb_src are strings)
Pytraj-based trajectory initialization (traj_src is pytraj.Trajectory)
Self-based trajectory initialization (traj_src is self)

All structures are regarded as trajectories (static PDB is view as a trajectory with only 1 frame).

Examples

>>> from nearl.io import Trajectory
>>> traj = Trajectory("traj.nc", "top.pdb")

Attributes:

top_filenamestr: The topology file of the trajectory
traj_filenamestr: The trajectory file of the trajectory
maskstr: The mask to select the atoms to be saved
mask_indicesnp.ndarray: The indices of the atoms selected by the mask
atomsnp.ndarray: The per-atom index of the trajectory
residuesnp.ndarray: The per-residue index of the trajectory

Methods

identity()	Return the identity of the trajectory
copy_traj()	Return a copy of the trajectory object
make_index()	Prepare the per-atom/per-residue index for the further trajectory processing
write_frame()	Save the trajectory to the file
add_dummy_points()	Add additional points to a frame for visual inspection

property identity#

Return the identity of the trajectory used for metadata retrieval.

Returns:

str: By default, it returns the trajectory file name

copy_traj()[source]#: Return a copy of the trajectory object

make_index()[source]#: Prepare the per-atom/per-residue index for the further trajectory processing;

write_frame(frames, outfile='', mask='', **kwarg)[source]#

Save the trajectory to the file

Parameters:

framesint, list_list or slice: The frame indices to be saved
outfilestr: The file name to save the trajectory
maskstr: The mask to select the atoms to be saved
kwargdict: Additional keyword arguments for the pytraj.save function

Examples

>>> from nearl.io import Trajectory
>>> traj = Trajectory("traj.nc", "top.pdb")
>>> traj.write_frame(0, "frame0.pdb", mask=":1-10")

add_dummy_points(coordinates, elements=None, frame_idx=0, outfile='')[source]#

Add additional points to a frame for visual inspection

Parameters:

coordinateslist: The list of coordinates to be added
elementsstr or list: The list of element symbols to be added
frame_idxint: The frame index to add the dummy points
outfilestr: The file name to save the new trajectory

class nearl.io.traj.MisatoTraj(pdbcode, misatodir, **kwarg)[source]#

Bases: Trajectory

Built-in implementation of the Misato HDF5 trajectory.

Original paper:

Siebenmorgen, T., Menezes, F., Benassou, S., Merdivan, E., Kesselheim, S., Piraud, M., Theis, F.J., Sattler, M. and Popowicz, G.M., 2023. MISATO-Machine learning dataset of protein-ligand complexes for structure-based drug discovery. bioRxiv, pp.2023-05.

Notes

The trajectory type has to be manually defined when pushing to the trajectory loader (see example). Solvents and ions are stripped for the alignment of the coordinates with the topology.

This module uses relative path to the misatodir to retrieve the trajectory. The following files are required to load the trajectory:

The topology file ({misatodir}/parameter_restart_files_MD/{pdbcode}/production.top.gz)
The trajectory file ({misatodir}/MD.hdf5)

Tip

Since there is no explicit annotation for the ligand part, we use a ligand indices map to extract the ligand part of the protein.

Examples

>>> from nearl.io import TrajectoryLoader, MisatoTraj
>>> trajs = [('5WIJ', '/MieT5/DataSets/misato_database/'), 
     ('4ZX0', '/MieT5/DataSets/misato_database/'), 
     ('3EOV', '/MieT5/DataSets/misato_database/'), 
     ('4K6W', '/MieT5/DataSets/misato_database/'), 
     ('1KTI', '/MieT5/DataSets/misato_database/')
    ]
>>> trajloader = TrajectoryLoader(trajs, trajtype=MisatoTraj)

Attributes:

pdbcodestr: The PDB code as the identity of the trajectory
topfilestr: The topology file of the trajectory
trajfilestr: The trajectory file of the trajectory

property identity#

Return the PDB code as the identity of the trajectory

Returns:

str: The PDB code of the trajectory

Trajectory

Contents

Trajectory#