read_dump command

Syntax:

read_dump file Nstep field1 field2 ... keyword values ... 

• file = name of dump file to read
• Nstep = snapshot timestep to read from file
• one or more fields may be appended

  field = x or y or z or vx or vy or vz or ix or iy or iz
    x,y,z = atom coordinates
    vx,vy,vz = velocity components
    ix,iy,iz = image flags in each dimension 
  

• zero or more keyword/value pairs may be appended
• keyword = box or replace or purge or trim or add or label or scaled or format

    box value = yes or no = replace simulation box with dump box
    replace value = yes or no = overwrite atoms with dump atoms
    purge value = yes or no = delete all atoms before adding dump atoms
    trim value = yes or no = trim atoms not in dump snapshot
    add value = yes or no = add new dump atoms to system
    label value = field column
      field = one of the listed fields or id or type
      column = label on corresponding column in dump file
    scaled value = yes or no = coords in dump file are scaled/unscaled
    format values = format of dump file, must be last keyword if used
      native = native LAMMPS dump file
      xyz = XYZ file
      molfile style path = VMD molfile plugin interface
        style = dcd or xyz or others supported my mofile
        path = optional path for location of molfile plugins 
  

Examples:

read_dump dump.file 5000 x y z
read_dump dump.file 5000 x y vx vy trim yes
read_dump ../run7/dump.file.gz 10000 x y z box yes 
read_dump dump.xyz 5 x y z box no format xyz
read_dump dump.xyz 10 x y z box no format molfile xyz ../plugins
read_dump dump.dcd 0 x y z format molfile dcd
read_dump dump.file 1000 x y z vx vy vz format molfile lammpstrj /usr/local/lib/vmd/plugins/LINUXAMD64/plugins/molfile 

Description:

Read atom information from a dump file to overwrite the current atom coordinates, and optionally the atom velocities and image flags and the simluation box dimensions. This is useful for restarting a run from a particular snapshot in a dump file. See the read_restart and read_data commands and the restart2data tool for alternative methods to do this. Also see the rerun command for a means of reading multiple snapshots from a dump file.

Note that a simulation box must already be defined before using the read_dump command. This can be done by the create_box, read_data, or read_restart commands. The read_dump command can reset the simulation box dimensions, as explained below.

Also note that reading per-atom information from a dump snapshot is limited to the atom coordinates, velocities and image flags, as explained below. Other atom properties, which may be necessary to run a valid simulation, such as atom charge, or bond topology information for a molecular system, are not read from (or even contained in) dump files. Thus this auxiliary information should be defined in the usual way, e.g. in a data file read in by a read_data command, before using the read_dump command, or by the set command, after the dump snapshot is read.

If the dump filename specified as file ends with ".gz", the dump file is read in gzipped format. You cannot (yet) read a dump file that was written in binary format with a ".bin" suffix, or to multiple files via the "%" option in the dump file name. See the dump command for details.

The format of the dump file is selected through the format keyword. If specified, it must be the last keyword used, since all remaining arguments are passed on to the dump reader. The native format is for native LAMMPS dump files, written with a "dump atom".html or dump custom command. The xyz format is for generic XYZ formatted dump files,

The molfile format supports reading data through using the VMD molfile plugin interface. This dump reader format is only available, if the USER-MOLFILE package has been installed when compiling LAMMPS.

The molfile format takes one or two additional values. The style value determines the file format to be used and can be any format that the molfile plugins support, such as DCD or XYZ. Note that DCD dump files can be written by LAMMPS via the dump dcd command. The path value specifies a list of directories which LAMMPS will search for the molfile plugins appropriate to the specified style. The syntax of the path value is like other search paths: it can contain multiple directories separated by a colon (or semi-colon on windows). The path keyword is optional and defaults to ".", i.e. the current directory.

Support for other dump format readers may be added in the future.

Global information is first read from the dump file, namely timestep and box information.

The dump file is scanned for a snapshot with a time stamp that matches the specified Nstep. This means the LAMMPS timestep the dump file snapshot was written on for the native format. However, the xyz and molfile formats do not store the timestep. For these formats, timesteps are numbered logically, in a sequential manner, starting from 0. Thus to access the 10th snapshot in an xyz or mofile formatted dump file, use Nstep = 9.

The dimensions of the simulation box for the selected snapshot are also read; see the box keyword discussion below. For the native format, an error is generated if the snapshot is for a triclinic box and the current simulation box is orthogonal or vice versa. A warning will be generated if the snapshot box boundary conditions (periodic, shrink-wrapped, etc) do not match the current simulation boundary conditions, but the boundary condition information in the snapshot is otherwise ignored. See the "boundary" command for more details.

For the xyz format, no information about the box is available, so you must set the box flag to no. See details below.

For the molfile format, reading simulation box information is typically supported, but the location of the simulation box origin is lost and no explicit information about periodicity or orthogonal/triclinic box shape is available. The USER-MOLFILE package makes a best effort to guess based on heuristics, but this may not always work perfectly.

Per-atom information from the dump file snapshot is then read from the dump file snapshot. This corresponds tot the specified fields listed in the read_dump command. It is an error to specify a z-dimension field, namely z, vz, or iz, for a 2d simulation.

For dump files in native format, each column of per-atom data has a text label listed in the file. A matching label for each field must appear, e.g. the label "vy" for the field vy. For the x, y, z fields any of the following labels are considered a match:

x, xs, xu, xsu for field x
y, ys, yu, ysu for field y
z, zs, zu, zsu for field z 

The meaning of xs (scaled), xu (unwrapped), and xsu (scaled and unwrapped) is explained on the dump command doc page. These labels are searched for in the list of column labels in the dump file, in order, until a match is found.

The dump file must also contain atom IDs, with a column label of "id".

If the add keyword is specified with a value of yes, as discussed below, the dump file must contain atom types, with a column label of "type".

If a column label in the dump file is not a match to a specified field, the label keyword can be used to specify which column label to associate with that field. An example is if a time-averaged coordinate is written to the dump file via the fix ave/atom command. The column will then have a label corresponding to the fix-ID rather than "x" or "xs". The label keyword can also be used to specify new column labels for fields id and type.

For dump files in xyz format, only the x, y, and z fields are supported. The dump file does not store atom IDs, so these are assigned consecutively to the atoms as they appear in the dump file, starting from 1. Thus you should insure that order of atoms is consistent from snapshot to snapshot in the the XYZ dump file. See the dump_modify sort command if the XYZ dump file was written by LAMMPS.

For dump files in molfile format, the x, y, z, vx, vy, and vz fields can be specified. However, not all molfile formats store velocities, or their respective plugins may not support reading of velocities. The molfile dump files do not store atom IDs, so these are assigned consecutively to the atoms as they appear in the dump file, starting from 1. Thus you should insure that order of atoms are consistent from snapshot to snapshot in the the molfile dump file. See the dump_modify sort command if the dump file was written by LAMMPS.

Information from the dump file snapshot is used to overwrite or replace properties of the current system. There are various options for how this is done, determined by the specified fields and optional keywords.

The timestep of the snapshot becomes the current timestep for the simulation. See the reset_timestep command if you wish to change this after the dump snapshot is read.

If the box keyword is specified with a yes value, then the current simulation box dimensions are replaced by the dump snapshot box dimensions. If the box keyword is specified with a no value, the current simulatoin box is unchanged.

If the purge keyword is specified with a yes value, then all current atoms in the system are deleted before any of the operations invoked by the replace, trim, or add keywords take place.

If the replace keyword is specified with a yes value, then atoms with IDs that are in both the current system and the dump snapshot have their properties overwritten by field values. If the replace keyword is specified with a no value, atoms with IDs that are in both the current system and the dump snapshot are not modified.

If the trim keyword is specified with a yes value, then atoms with IDs that are in the current system but not in the dump snapshot are deleted. These atoms are unaffected if the trim keyword is specified with a no value.

If the add keyword is specified with a yes value, then atoms with IDs that are in the dump snapshot, but not in the current system are added to the system. These dump atoms are ignored if the add keyword is specified with a no value.

Note that atoms added via the add keyword will have only the attributes read from the dump file due to the field arguments. If x or y or z is not specified as a field, a value of 0.0 is used for added atoms. Added atoms must have an atom type, so this value must appear in the dump file.

Any other attributes (e.g. charge or particle diameter for spherical particles) will be set to default values, the same as if the create_atoms command were used.

Note that atom IDs are not preserved for new dump snapshot atoms added via the add keyword. The procedure for assigning new atom IDS to added atoms is the same as is described for the create_atoms command.

Atom coordinates read from the dump file are converted into absolute, unscaled coordinates, relative to the box dimensions of the snapshot. These coordinates may then be assigned to an existing or new atom in the current simulation. The coordinates will be remapped to the simulation box, whether it is the original box or the dump snapshot box. If periodic boundary conditiona apply, this means the atom will be remapped back into the box if necessary. If shrink-wrap boundary conditions apply, the new coordinates may change the current box dimensions. If fixed boundary conditions apply, the atom will be lost if it is outside the simulation box.

For native format dump files, the 3 xyz image flags for an atom in the dump file are set to the corresponding values appearing in the dump file if the ix, iy, iz fields are specified. If not specified, the image flags for replaced atoms are not changed and image flags for new atoms are set to default values. The remapping procedure described in the previous paragraph can change images flags for all atoms (old and new) if periodic boundary conditions are applied to remap an atom back into the simulation box. Note that inconsistent image flag values can result if you use image flag fields from the dump file but do not also use the dump file box parameters.

LAMMPS knows how to compute absolute, unscaled coordinates for the snapshot column labels discussed above, e.g. x, xs, xu, xsu. If another column label is assigned to the x or y or z field via the label keyword, e.g. for coordinates output by the fix ave/atom command, then LAMMPS needs to know whether the coordinate information in the dump file is scaled or unscaled. This can be set via the scaled keyword. The value of the scaled keyword is ignored for field x or y or z if the label keyword is not used to assign a column label to that field.

The scaled vs unscaled setting must be consistent for any of the x, y, z fields that are specified. If the dump file coordinates are scaled and the simulation box is triclinic, then all 3 of the x, y, z fields must be specified, since they are all needed to generate absolute, unscaled coordinates.

Restrictions:

To read gzipped dump files, you must compile LAMMPS with the -DLAMMPS_GZIP option - see the Making LAMMPS section of the documentation.

The molfile dump file formats are part of the USER-MOLFILE package. They are only enabled if LAMMPS was built with that packages. See the Making LAMMPS section for more info.

Related commands:

dump, dump molfile, read_data, read_restart, rerun

Default:

The option defaults are box = yes, replace = yes, purge = no, trim = no, add = no, scaled = no, and format = native.