load 
(v. 13.00)

The load command by itself or with empty quotes reloads the current file.

An optional keyword, APPEND, DATA, FILES, MENU, MODELS, or TRAJECTORY, may be supplied prior to the quoted filename. Other keywords are ignored. (Jmol does not use the Chime-style keyword to specify "file format". Rather, file format can be forced by prefixing a filename with "xxxx::" where "xxxx" is a Jmol file type. However, this should not be necessary in most cases, since Jmol determines file type by scanning the first lines of a file for file-type-specific content. (In certain cases, where there are extensive comments at the beginning of a file, it is possible for Jmol to fail to discover the file type or to misassign it. In that case, xxxx:: should be used.)

Jmol automatically determines file type based upon the contents of the file. Quotes are recommended. Files containing fractional coordinates are displayed with their unit cell visible. load "" reloads the current file. For files containing multiple models, an optional integer after the file name will load only the specified model. If this number is negative, it refers to a specific vibrational mode in files that contain that information . If this number is 0, it refers to the last model in the set . In the case of PDB and mmCIF files, this number refers to the model number specied in the MODEL record or _atom_site.pdbx_PDB_model_num field, respectively.

An additional way to load a specific set of models involes using an array just after the file name, starting with model 1 for the first. The order of models is not significant. Thus, [1 2 3] is the same as [3 2 1] and loads the first three models only. (Note that if you want to display models in a different sequence then they appear in a file, you can do that with the frame command.) In the case of PDB and mmCIF files, these numbers refer to the model number specied in the MODEL record or _atom_site.pdbx_PDB_model_num field, respectively. A comparison of load options using just a model number or an array is given below:

load ""	loads the first model in a file when not PDB or mmCIF or the model with MODEL 1 record for a PDB file or the model with _atom_site.pdbx_PDB_model_num = 1 for an mmCIF file.
load "" [1]	same as load "" 1; brackets allow for more than one model.
load MODELS ({1}) ""	(see load MODELS) always loads the SECOND model in a file, regardless of its indicated model number.

File format can be forced by prefixing a filename with "xxxx::" where "xxxx" is a Jmol file type. This should not be necessary in most cases, since Jmol determines file type by scanning the first lines of a file for file-type-specific content. In certain cases, however, where there are extensive comments at the beginning of a file, or a file type (such as MDCRD) does not have any distinguishing characteristics, it is possible for Jmol to fail to discover the file type or to misassign it. In that case, adding "xxxx::" is necessary.

Loads a remote file and then also saves it locally. For use with the Jmol application and signed applet only.

Loads the file with the name specified by variable variableName. You can load a set of files that are defined in an array variable.

Loads the file data given in quotes. Generally this would be a small molecule in JME or XYZ format.

You can load a model from data contained in a variable. This allows modification of the data prior to display, for example. Quotes are necessary, as without them -- load @x -- it is the file name that is expected in variable x, not the file contents. For example: x = load("quartz.cif");load "@x" {2 2 2};reset x. Note that to save memory, it is a good idea to clear the variable using reset x just after loading. An interesting aspect of this option is that it allows data from a remote file to be saved in the state. This means that if you use write state "somefile.spt", then that single state script will contain the full file data in a DATA statement. It will be "transportable" and no additional model file will be required. (See also the PNGJ format for the write command, which is even better.)

Loads the file data given in variable x; same as load "@x" but in the same syntax as write VAR x.

You can load mmCIF files directly from http://www.ebi.ac.uk/pdbe. Simply preface the four-letter PDB id code with "*", and Jmol will load the file. Using load "*?" offers the user a prompt for a PDB id. As for other file laoding, you can adding "AS ." ("as" with a period) you can save that file automatically in the default directory (as xxxx.cif), and using, for example, load *1crn AS "myfile.cif", you can save it to some other local file name. Starting with Jmol 14.4, you can add "/dssr" to the PDB id to also read RNA/DNA secondary structure information from Columbia University, "/rna3d" for RNA secondary structure annotations from http://rna.bgsu.edu/rna3dhub, sequence domain annotations using "/dom", and structure validation annotations using "/val".

Appending a * to a PDBe ID loads a PDBe "updated" CIF file, which allows PDB CONECT-like bond creation. An example of such a file can be found here. This is a suitable replacement for PDB CONECT, integrating information found in _chem_comp_bond and _struct_conn categories. Note that the presence of _chem_comp_bond records in these files will enable processing of _struct_conn as well, regardless of filter "addbonds".

You can load PDB files directly from http://www.rcsb.org or another server of your choice. (This depends upon the setting of set loadFormat). Simply preface the four-letter PDB id code with "=", and Jmol will load the file. Using load "=?" offers the user a prompt for a PDB id. Adding "AS ." ("as" with a period) you can save that file automatically in the default directory (as xxxx.pdb.gz), and using, for example, load =1crn AS "myfile.pdb", you can save it to some other local file name. The default is to transfer the file in g-zipped format; If you add ".pdb" to the expression -- load =1crn.pdb, for example -- then Jmol will transfer and save the uncompressed PDB file. Starting with Jmol 14.2, you can add "/dssr" to the PDB id to also read RNA/DNA secondary structure information from Columbia University. Starting with Jmol 14.4, you can load "/rna3d" for RNA secondary structure annotations from http://rna.bgsu.edu/rna3dhub, sequence domain annotations using "/dom", and structure validation annotations using "/val".

You can load PDB ligand (chemical component) files directly from http://www.rcsb.org or another server of your choice. (This depends upon the setting of set loadLigandFormat). Simply preface the three-letter PDB id code with "==", and Jmol will load the file. Using load "==?" offers the user a prompt.

You can load files from a named database. Options (Jmol.4.6) include:

mp	load =mp/24972	loads a file from http://www.materialsproject.org
ligand	load =ligand/hem	RCSB chemical component file
nci	load =nci/CC	NCI CACTVS Resolver
pdb	load =pdb/1crn	RCSB pdb files
iucr	load =iucr/wf5113sup1	IUCr Acta Cryst B supplemental files
cod	load =cod/1000002	Crystallography Open Database
iucr	load =pdbe/1crn	CIF file from PDBe
iucr	load =pdbe2/1crn	CIF file from PDBe (alternative)
aflow	load =aflow/AgAu	AFLOW binary metal alloy file
magndata	load =magndata/1.1.3	MAGNDATA magnetic crystal database (Bilbao Crystallographic Server
ams	load =ams/quartz	American Mineralogist Crystal Structure Database

You can load models from PubChem. Preface the compound name with colon. Quotes are only required if nonalphanumeric characters are present in the name.

You can load SMILES strings, and Jmol will turn them into 3D models using the NIH Cactus server. As for reading files from any source outside your domain, you will have to use the signed applet or Jmol application to do this. These files can be saved as MOL files using write xxx.mol or load $xxxx AS "myfile.mol", and if the conformation is not to your liking, switching to set modelkitMode or using set picking dragMinimize you can quickly adjust the model to the desired conformation. Quotation marks should be used for names that include the space character: load "$ethyl acetate".

The service at https://cactus.nci.nih.gov is not limited to SMILES strings. The database includes several million known compounds, which are accessible by CAS registry number, InChI codes, and chemical name as well. For example, load $57-88-5 loads a model of chlolesterol, and load $taxol as "taxol.mol" loads a model of taxol and saves it on your hard drive. This service taps into the OPSIN service of Cambridge University to return structures from IUPAC or IUPAC-like names.

Loads structure from NCI/CADD with prompt.

Loads chemical component from RCSB with prompt.

Loads PDB ID from RCSB with prompt.

Loads PDB ID from EBI with prompt.

Loads structure from pubChem with prompt.

Using just LOAD $ will load the current structure from NCI/CADD by passing a SMILES string to that service. The NCI/CADD service will create 3D models directly from SMILES using CORINA, if necessary. Using load "$?" offers the user a prompt.

Using just LOAD : will load the current structure from PubChem by passing a SMILES string to that service. Note that PubChem, unlike NCI/CADD, will only deliver models that have been deposited in its database. Using load ":?" offers the user a prompt.

chEBI 2D molecule load, with minimal 100-step minimization.

Use for molecular crystal structures and loads one or more unit cells (as specified by {nx ny nz})
exploring covalent bonding so that molecules that have their centroid (center of geometry) within the specified block of cells are built. For example:
load =cod/1001253 {1 1 1} centroid

Note that {1 1 1} here is unnecessary, as it is the default.

Allows prefix "Hall:" to specify Hall notation along with SPACEGROOUP 3 or SPACEGROUP "10:b". Allows for experimentation with different space groups.

Creates an empty structure with the designated space group and unit cell; User is expected to match the correct type of unit cell to the given space group.

The ORIENTATION keyword specifies that after the file is loaded, the orientation should be returned to its original value. This orientation can be restored later using restore orientation PRELOAD.

An alternative to the $ syntax for loading SMILES strings.

The FILL option, described more fully under Crystallographic Options, allows filling an arbitrary space with a set of atoms from a crystallographic file. When the file has no unit cell, this option simply loads the file with the specified boundbox.

For individual file types, it is possible to filter the data in the file as it is loaded. The FILTER keyword followed by a quoted string allows for this. The FILTER keyword must be the last keyword in the LOAD command. Specific filters include:

General file types
all	FILTER "NAME=..."	loads only those models with a name that contains a specified set of characters.
all	FILTER "CENTER"	centers all models in a file based on the first model"s position.
all	FILTER "REVERSEMODELS"	delivers the models in reverse order . Useful for intrinsic reaction coordinate (IRC) animations.
2D data	FILTER "NOH" or FILTER "noHydrogen"	For 2D structures (MOL, V3000, CDX, CDXML, JME), loads the structure as is, without adding H or converting to 3D. This may be useful for automated processes that simply need to create SMILES or InChI but not actually display any structure.
2D data	FILTER "NOMIN"	For 2D structures (MOL, V3000, CDX, CDXML, JME), loads the structure with added hydrogens but without converting to 3D. Atoms may be adjusted in preparation for minimization, but no minimization is carried out.
many	FILTER "NOVIB"	skips the reading of vibrations for any file that contains them.
Small molecule file types
CDX, CDXML	FILTER "NO3D"	Some CDX and CDXML files contain both 2D and 3D coordinates. The default for Jmol is to read the 3D coordinates if they are present. This is not generally advisable, however, as they can be quite odd, particularly if abbreviations such as Ph or t-Bu have been used. This setting forces any 3D coordinates in the file to be ignored, loading the model in 2D followed by conversion to 3D in Jmol. The combination filter "no3d noH nomin" will display the CDX or CDXML file as it would be drawn, using only the 2D coordinates, adding no hydrogen atoms, and doing no minimization.
MOL	FILTER "2D"	indicates to consider the file a 2D file and to apply a automatic hydrogen addition and 2D-to-3D conversion immediately upon loading. "2D-noMIN" does the hydrogen addition but no minimization.
MOL	FILTER “no 3D”	Loads a 2D model with no conversion to 3D, particularly for V3000 and ChemDraw; no effect for standard 3D models.
MOL	FILTER “noHydrogen”(Or just “NOH”)	Loads a 2D model with no additional hydrogens and no minimization; allows for SMILES and InChI creation directly from the 2D model.
Biomolecular file types
PDB, mmCIF, MMTF, binCIF	FILTER "ALLHET"	By itself, does nothing, but in conjunction with any other atom filter, such as /=2 or *.CA passes any HETATM regardless of that atom filter.
mmCIF	FILTER "ADDBONDS"	One of the deficiencies of of the mmCIF format as originally defined is that it does not reproduce the CONECT records of PDB files. The ADDBONDS filter instructs Jmol to process mmCIF_struct_conn records. These records are similar to PDB CONECT records, but they indicate only inter-group connections, connecting one group with another. Types included by Jmol include covale, covale_base, covale_phosphate, covale_sugar, disulf, and metalc. Ignored types include hydrog, mismat, modres, and saltbr. This filter is not necessary when reading updated CIF files from PDBe (See *xxxx*, above).
mmCIF, PDB, MMTF	FILTER "ADDHYDROGENS"	Add hydrogens regardless of the setting of pdbAddHydrogens.
mmCIF, PDB, MMTF	FILTER "ASSEMBLY n"	loads a specific CIF assembly. (Same as BIOMOLECULE.) For example, load =1vif.cif filter "ASSEMBLY 1". Specific label_asym_id can be selected as well: load =1vif.cif filter "ASSEMBLY 1;$A" (just label_asym_id A); load =1vif.cif filter "ASSEMBLY 1;!$C" (just label_asym_id values not C).
mmCIF, PDB, MMTF	FILTER "ATOM"	selects only ATOM records (see also FILTER "HETATM")
mmCIF, PDB, MMTF	FILTER "BIOMOLECULE n"	where n is a number > 0 indicating which biomolecule to load. In addition, #<n or any number of # n or !#n can be indicated in order to load just a specific subset of biomolecular transformations related to the specified biomolecule. Adding "NOSYMMETRY" along with BIOMOLECULE, indicates that the symmetry transformations in the REMARK 350 BIOMT records should be ignored. See also the BMCHAINS filter option.
mmCIF, PDB, MMTF	FILTER "BMCHAINS=mode"	Rename chains when applying symmetry operations so that they have distinct names. "BMCHAINS=0" appends the symmetry operation number (2,3,...) to chain identifiers of all generated atoms, so {A B A B A B} mitght become {A B A2 B2 A3 B3}. "BMCHAINS=1" extends the single-character alphabetization, so {A B A B A B} might become {A B C D E F}.
mmCIF, PDB, MMTF	FILTER "BYCHAIN"	designed for extremely large high-symmetry biological assemblies, this filter loads a file in a coarse-grained fashion, assigning one pseudoatom per chain. The size of this atom can be made larger than the atom-size maximum of 16 Angstroms using set particleRadius with a larger value and then setting spacefill 30 (or any number greater than 16). Normally accompanied by BIOMOLECULE or ASSEMBLY, for example: load =3j3q.cif filter "ASSEMBLY 1;BYCHAIN".
mmCIF, PDB, MMTF	FILTER "BYSYMOP"	similar to BYCHAIN, but one pseudoatom per symmetry operation. For example: load =1m4x.cif filter "ASSEMBLY 1;BYSYMOP".
mmCIF, PDB, MMTF	FILTER "HETATM"	selects only HETATM records (see also FILTER "ATOM")
mmCIF, PDB, MMTF, GROMACS, MDTOP	FILTER "[XXX], .XXX, :X, %X"	to specify inclusion or exclusion of specific residue types, atom types, chains, or alternative locations. The prefix "!" indicates NOT. Within each of these four sets, multiple selections are treated as "OR" without "!" and "AND" when "!" is present. With % (alternative location), atoms with no alternative location marking are ALWAYS loaded regardless of this filter. For example, load "1sva.pdb" filter "*.CA,%B" loads only alpha carbons that correspond to the conformation involving alternative location B; filter "![HOH]" filters out water molecules.
PDB	FILTER "CONF n"	Loads only a specific configuration based on alternative locations. "CONF 1" loads only atoms in the first configuration -- usually %A and % ("undesignated")
PDB	FILTER "TYPE i,n"	loads customized column data into the atomType property of the atoms. The two numeric parameters are 1-based column number and number of columns. For example: load xxx.pdb filter "TYPE 73,4".
PDB	FILTER "TYPE i,n=XXX"	loads customized column data into the atomType property of the atoms, selecting only those atoms starting with the specified characters in the specified field.
P2N	FILTER "ALTNAME"	Use the altName field of the P2N file for atom names.
PYMOL	FILTER "DOCACHE"	caches surface data, creating a JVXL equivalent, specifically useful for JSmol. When using the "doCACHE" option (in the Jmol Java application), file loading should be followed immediately with WRITE PNGJ "xxx.PNG". The PNGJ file can then be loaded much more efficiently than the PSE file on a web page using JSmol/HTML5.
PYMOL	FILTER "NOSURFACE"	load a PyMOL file without surface creation, which can be slow.
PyMOL	FILTER "ShowValence"	May be needed for PyMOL files that have not been saved with valence setting ON in order to show multiple bonds in Jmol. Note that the PyMOL valence_mode setting is NOT implemented in Jmol.
Crystallographic file types
CIF, various	FILTER "CELL=abc;offset"	specifies a specific cell in relation to the unit cell given in the file using MCIF parent/standard cell notation. For example, filter "CELL=a/2,2b,c;0,0,1/2" (see unitcell).
CIF, various	FILTER "LATTICESCALING=1.2"	Scales the model based on a crystal lattice expansion factor, allowing direct comparison of related structures with slightly different lattice scalings.
CIF, various	FILTER "POLYMERX"	Loads a crystal structure as a linear polymer, packing only in the X direction. Compatible only with P1 symmetry. For example: load =ams/quartz 1 packed;load inline @{write("CIFP1")} {3 1 1} filter "POLYMERXY"
CIF, various	FILTER "Precision=n" and FILTER "lowPrecision"	"PRECISION=n" where n is up to 16 digits specifies the precision to use for a given file, possibly lowering the precision from what is in the file. For example, AFLOWLIB files tend to have their precision hidden by padding with zeros: U1 U 4 a 0.00000 0.00000 0.00000 1.00000 B1 B 48 i 0.50000 0.66600 0.66600 1.00000 This could cause trouble with some edge cases or when determining Wyckoff positions. One could prevent the problem using filter "PRECISION=3" in this case, explicitly ignoring the padding and turning "0.666" into "0.66666667". Using filter "lowPrecision", one can set the precision to 4 (+/-0.0001) for fractional coordinates. This filter specifies that crystal structure readers should not attempt to raise the precision to single- or double-precision even if the file contains double-precision data. It also sets PACKED 0.0001 rather than the standard default value of 0.02 (which tends to allow too many atoms to be included) in some cases.
CIF, various	FILTER “SLABXY” or “POLYMERX”	Loads a crystal structure as a slab (SLABXXY) or polymer (POLYMERX). Packs only in the periodic direction (xy or x). This option is compatible only with P1 symmetry. For example: load =ams/quartz 1 packed x = write("CIFP1") load INLINE @x {3 3 1} filter "SLABXY"
CIF, various	FILTER "SYMOP=n"	loads just the atoms that are the result of the nth symmetry operator.
CIF	FILTER "MOLECULAR"	instructs Jmol to attempt to complete molecules using either the bonding information in the CIF file or standard van der Waals radii. If this filter is not present and no lattice is indicated either, if bonding information is present in the file, Jmol will use that information to automatically complete molecules.
CIF	FILTER "NOIDEAL"	for mmCIF chemical composition files (for example, load ==HEM) uses the _chem_comp_atom.pdbx_model_Cartn_x set of coordinates rather than the default _chem_comp_atom.pdbx_model_Cartn_x_ideal set
CIF	FILTER "NOWYCKOFF"	Skip reading atom_site_wyckoff_label, allowing Jmol to calculate these positions itself. (For Jmol development testing purposes, primarily).
CIF	filter "StopOnSHELXHKL"	Improves load time by stopping reading when a _shelx_hkl_file tag is found. This filter should not be used with multiple structures. Prior to Jmol 16.2.19, this filter was effectively set, but that causes problems for multiple-structure CIF files.
CIF, JANA2006	FILTER "NOSYM"	do not apply symmetry
mCIF	FILTER "CELL=parent"	loads and packs the parent cell, if designated in the file. If packing is not desired, use "nopackcell" instead of "cell".
mCIF	FILTER "cell=standard"	loads and packs the standard cell, if designated in the file. If packing is not desired, use "nopackcell" instead of "cell".
msCIF, JANA2006	FILTER "MODAVERAGE"	do not read modulation
msCIF, JANA2006	FILTER "MODAXES=xyz"	read modulation only along specific axes -- x, y, and/or z
msCIF, JANA2006	FILTER "MODCELL=n"	use subsystem n for initially created unit cell, generating the command unitcell {%1} for instance upon file loading.
msCIF, JANA2006	FILTER "MODNOPACK"	do not pack subsystem cells
msCIF, JANA2006	FILTER "MODT=t"	set initial (zero-point) modulation to t, where t can be a decimal number or a fraction such as 1/3.
msCIF, JANA2006	FILTER "MODTUV=t,u,v"	set initial (zero-point) modulations for up to three cell wave vectors, where t,u,v can be decimal numbers or fractions such as 1/3.
BCS	FILTER "NONORM"	reads displacement data from BCS files (particularly in relation to AMPLIMODES files) without normalization, allowing comparison of displacement magnitudes between different models.
CGD	FILTER "NOBONDSYM"	Do not apply symmetry operators to bonds -- just create the ones present in the file.
CGD	FILTER "NOPACK"	Do not pack cell (default for this type is to pack the cell).
topoCIF	FILTER "TOPOL" FILTER "TOPOLNET"	For topological CIF files, reads the net from the _topol_* data blocks. Displays net 1 by default. load "xxx" {1 1 1} filter "TOPOL" will do the same, constructing the {1 1 1} unit cell, with symmetry. Also works with fill or packed. Displays net 1 by default. load "xxx" {1 1 1} filter "TOPOLNet=2" displays net.id=2; load "xxx" {1 1 1} filter "TOPOLNet=Net_2" displays net.name="Net_2". Additional filters include filter "topos_type=x" where x is v, vw, hb, or some concatenation of those using "+", and filter "topos_ignore", which skips reading of _topol records, reading as a standard CIF file.
Quantum computing file types
many	FILTER "NOMO"	skips the reading of molecular orbitals for any file that contains them.
CASTEP	FILTER "Q=..."	for .phonon files, filter "q=n" loads the nth q-point frequency modes. "q=all" reads all phonon modes. For files with multiple extrapolations to q=0, "q=0.1", "q=0.2", etc... loads a specific extrapolation. Q-points can also be specified in terms of fractions, "q=(1/2 1/2 1/2)", or decimals, "q=(0.333 0.111 0.333)". If no filter is specified, but the SUPERCELL keyword is used, Jmol will attempt to find the appropriate matching q-point data for that supercell.
CRYSTAL	FILTER "CONV"	load conventional, not primitive cells.
CRYSTAL	FILTER "INPUT"	load input coordinates only.
CSF, SPARTAN	FILTER "NOORIENT"	prevents application of the rotation matrix found in the file as the default rotation.
GAMESS, GAUSSIAN, GenNBO, Jaguar, Molden, NWChem, PSI, and QCHEM	FILTER "xxx"	where "xxx" is a word on the line Jmol uses to identify a molecular orbital. This allows selective loading of specific types of molecular orbitals -- such as "alpha", "beta", or "NBO" -- for any of these file types, "POPULATION" or "EDMISTON" or "PIPEK" for GAMESS. ("NBO" refers to orbitals generated with the NBO 5.0 option AONBO=P; "EIGEN" or "!NBO" will skip loading of these orbitals.)
GAMESS, Gaussian, QChem, NWChem	FILTER "NBOCHARGES"	specifies to use atomic partial charges from the NBO calculation. Note that this flag is independent of the NBO filter option. To use both, separate them with a comma: load "myfile.out" filter "NBO,nboCharges".
GAMESS (US)	CHARGE=LOW	indicates to load Lowden charges rather than Mulliken charges.
Gaussian	FILTER "ORIENTATION:..."	Orientation filters returning the specified orientation, one of filter "orientation:input" (includes Z-Matrix), filter "orientation:standard", filter "orientation:all". Note that vibrations are only returned if orientation is ALL or STANDARD.
Molden	FILTER "NOSORT"	Do not sort MOs by energy.
Molden	FILTER "OPTONLY"	Load optimized structure only.
Molden	FILTER "SYM=xxx"	Load only molecular orbitals for which the Sym= line in the [MO] block contains the text "xx".
Molden	FILTER "VIBONLY"	Load vibrations only, not the base geometry.
SPARTAN	FILTER "MULLIKEN"	loads Mulliken charges rather than the default ESP charges. (ESP charges are better for displaying molecular electrostatic potential maps using isosourface molecular map MEP.)

See also:

undefined