ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'master' into sdpd to resolve merge conflicts

Browse Source 
# Conflicts:
#	cmake/CMakeLists.txt
#	src/Makefile
This commit is contained in:
Axel Kohlmeyer
2018-11-08 16:32:02 -05:00
parent bbb371134d 3681bc853d
commit e0955f6434
71 changed files with 9459 additions and 528 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
doc/src/Build_extras.txt
View File

@ -41,6 +41,7 @@ This is the list of packages that may require additional steps.
"USER-ATC"_#user-atc,
"USER-AWPMD"_#user-awpmd,
"USER-COLVARS"_#user-colvars,
"USER-PLUMED" _#user-plumed,
"USER-H5MD"_#user-h5md,
"USER-INTEL"_#user-intel,
"USER-MOLFILE"_#user-molfile,
@ -712,6 +713,62 @@ a corresponding Makefile.lammps.machine file.
:line
USER-PLUMED package :h4,link(user-plumed)
[CMake build]:
[Traditional make]:
Before building LAMMPS with this package, you must first build
PLUMED. We recommending building PLUMED separately to LAMMPS using
the instructions that can be found at http://plumed.github.io/doc-master/user-doc/html/_installation.html.
Before compiling LAMMPS you can then install the fix plumed command
and compile LAMMPS in the usual manner:
make yes-user-plumed
make machine :pre
Once this compilation completes you should be able to run LAMMPS in the usual
way. When running LAMMPS with an input script that contains a fix
plumed command LAMMPS will try to call the PLUMED runtime library. PLUMED
must therefore be available in your path if LAMMPS is compiled in this way.
On some machines it is not possible to call runtime libraries in the way described
above. When compiling on these machines it is thus better to statically link
PLUMED when compiling LAMMPS. To do this you must either download a PLUMED
tarball from http://www.plumed.org/get-it or clone it using
git clone https://github.com/plumed/plumed2.git. If you download the tarball
unpack it in the /lib/plumed directory. Similarly if you clone
it clone it to the /lib/plumed directory as if there is a version of PLUMED within
this directory LAMMPS will always try to statically link the version of PLUMED
that this directory contains instead of dynamically linking the library.
Once you have downloaded PLUMED into /lib/plumed you must again build the code
here by following the instructions that can be found at
http://plumed.github.io/doc-master/user-doc/html/_installation.html.
You can statically link PLUMED manually and if you want to access the full
range of PLUMED functionalities this is what you should do. If you only want the
basic range of functionalities, however, (i.e. no user contributed modules) then
you can download and compile PLUMED in one step from the lammps/src dir, using a
command like like those below:
make lib-plumed # print help message
make lib-plumed args="-b" # download and build the latest stable version of PLUMED
These commands will simply invoke the lib/plumed/Install.py script with
args specified. Furthermore, once the script has completed you should
have a compiled version of PLUMED. With this built you can install/un-install
PLUMED and build LAMMPS in the usual manner:
make yes-user-plumed
make machine :pre
make no-user-plumed
make machine :pre
:line
USER-H5MD package :h4,link(user-h5md)
To build with this package you must have the HDF5 software package

BIN
doc/src/PDF/colvars-refman-lammps.pdf
View File

Binary file not shown.

28
doc/src/Packages_details.txt
View File

@ -1201,6 +1201,34 @@ examples/USER/colvars :ul
:line
USER-PLUMED package :link(USER-PLUMED),h4
[Contents:]
The fix plumed command allows you to use the plugin for molecular
dynamics PLUMED to analyse and bias your LAMMPS trajectory on the fly.
In practise PLUMED is called from within the lammps input script by using
the "fix plumed _fix_plumed.html command.
[Authors:] The PLUMED library is written and maintained by
Massimilliano Bonomi, Giovanni Bussi, Carlo Camiloni and
Gareth Tribello.
[Install:]
This package has "specific installation
instructions"_Build_extras.html#gpu on the "Build
extras"_Build_extras.html doc page.
[Supporting info:]
src/USER-PLUMED/README
lib/plumed/README
"fix plumed "_fix_plumed.html
examples/USER/plumed :ul
:line
USER-DIFFRACTION package :link(PKG-USER-DIFFRACTION),h4
[Contents:]

4
doc/src/fix_bond_break.txt
View File

@ -137,8 +137,8 @@ doc page for more info.
[Related commands:]
"fix bond/create"_fix_bond_create.html, "fix
bond/swap"_fix_bond_swap.html, "dump local"_dump.html,
"special_bonds"_special_bonds.html
bond/react"_fix_bond_react.html, "fix bond/swap"_fix_bond_swap.html,
"dump local"_dump.html, "special_bonds"_special_bonds.html
[Default:]

4
doc/src/fix_bond_create.txt
View File

@ -232,8 +232,8 @@ doc page for more info.
[Related commands:]
"fix bond/break"_fix_bond_break.html, "fix
bond/swap"_fix_bond_swap.html, "dump local"_dump.html,
"special_bonds"_special_bonds.html
bond/react"_fix_bond_react.html, "fix bond/swap"_fix_bond_swap.html,
"dump local"_dump.html, "special_bonds"_special_bonds.html
[Default:]

95
doc/src/fix_bond_react.txt
View File

@ -24,11 +24,11 @@ common_keyword = {stabilization} :l
{stabilization} values = {no} or {yes} {group-ID} {xmax}
{no} = no reaction site stabilization
{yes} = perform reaction site stabilization
{group-ID} = user-assigned ID for all non-reacting atoms (group created internally)
{group-ID} = user-assigned prefix for the dynamic group of non-reacting atoms
{xmax} = xmax value that is used by an internally created "nve/limit"_fix_nve_limit.html integrator :pre
react = mandatory argument indicating new reaction specification :l
react-ID = user-assigned name for the reaction :l
react-group-ID = only atoms in this group are available for the reaction :l
react-group-ID = only atoms in this group are considered for the reaction :l
Nevery = attempt reaction every this many steps :l
Rmin = bonding pair atoms must be separated by more than Rmin to initiate reaction (distance units) :l
Rmax = bonding pair atoms must be separated by less than Rmax to initiate reaction (distance units) :l
@ -41,14 +41,18 @@ react = mandatory argument indicating new reaction specification :l
fraction = initiate reaction with this probability if otherwise eligible
seed = random number seed (positive integer)
{stabilize_steps} value = timesteps
timesteps = number of timesteps to apply internally created nve/limit.html :pre
timesteps = number of timesteps to apply internally created nve/limit.html
{update_edges} value = {none} or {charges} :l
none = do not update topology near the edges of reaction templates
charges = update atomic charges of all atoms in reaction templates
custom = force the update of user-specified atomic charges :pre
:ule
[Examples:]
molecule mol1 pre_reacted_topology.txt
molecule mol2 post_reacted_topology.txt
fix 5 all bond/react stabilization no react myrxn1 all 1 0 3.25 mol1 mol2 map_file.txt :pre
fix 5 all bond/react react myrxn1 all 1 0 3.25 mol1 mol2 map_file.txt :pre
molecule mol1 pre_reacted_rxn1.txt
molecule mol2 post_reacted_rxn1.txt
@ -57,7 +61,7 @@ molecule mol4 post_reacted_rxn2.txt
fix 5 all bond/react stabilization yes nvt_grp .03 &
react myrxn1 all 1 0 3.25 mol1 mol2 map_file_rxn1.txt prob 0.50 12345 &
react myrxn2 all 1 0 2.75 mol3 mol4 map_file_rxn2.txt prob 0.25 12345
fix 6 nvt_grp nvt temp 300 300 100 # set thermostat after bond/react :pre
fix 6 nvt_grp_REACT nvt temp 300 300 100 # set thermostat after bond/react :pre
[Description:]
@ -99,19 +103,29 @@ involved in any new reactions. The {xmax} value keyword should
typically be set to the maximum distance that non-reacting atoms move
during the simulation.
The group-ID set using the {stabilization} keyword should be a
previously unused group-ID. It cannot be specified as 'all'. The fix
bond/react command creates a "dynamic group"_group.html of this name
that includes all non-reacting atoms. This dynamic group-ID should
then be used by a subsequent system-wide time integrator such as nvt,
npt, or nve, as shown in the second example above. It is currently
necessary to place the time integration command after the fix
bond/react command due to the internal dynamic grouping performed by
fix bond/react.
The group-ID set using the {stabilization} keyword can be an existing
static group or a previously-unused group-ID. It cannot be specified
as 'all'. If the group-ID is previously unused, the fix bond/react
command creates a "dynamic group"_group.html that is initialized to
include all atoms. If the group-ID is that of an existing static
group, the group is used as the parent group of new,
internally-created dynamic group. In both cases, this new dynamic
group is named by appending '_REACT' to the group-ID, e.g.
nvt_grp_REACT. By specifying an existing group, you may thermostat
constant-topology parts of your system separately. The dynamic group
contains only non-reacting atoms at a given timestep, and therefore
should be used by a subsequent system-wide time integrator such as
nvt, npt, or nve, as shown in the second example above. The time
integration command should be placed after the fix bond/react command
due to the internal dynamic grouping performed by fix bond/react.
NOTE: The internally created group currently applies to all atoms in
the system, i.e. you should generally not have a separate thermostat
which acts on the 'all' group.
NOTE: If the group-ID is an existing static group, react-group-IDs
should also be specified as this static group, or a subset.
NOTE: If the group-ID is previously unused, the internally created
group applies to all atoms in the system, i.e. you should generally
not have a separate thermostat which acts on the 'all' group, or any
other group.
The following comments pertain to each {react} argument:
@ -155,7 +169,17 @@ Some atoms in the pre-reacted template that are not reacting may have
missing topology with respect to the simulation. For example, the
pre-reacted template may contain an atom that would connect to the
rest of a long polymer chain. These are referred to as edge atoms, and
are also specified in the map file.
are also specified in the map file. When the pre-reaction template
contains edge atoms, not all atoms, bonds, charges, etc. specified in
the reaction templates will be updated. Specifically, topology that
involves only atoms that are 'too near' to template edges will not be
updated. The definition of 'too near the edge' depends on which
interactions are defined in the simulation. If the simulation has
defined dihedrals, atoms within two bonds of edge atoms are considered
'too near the edge.' If the simulation defines angles, but not
dihedrals, atoms within one bond of edge atoms are considered 'too
near the edge.' If just bonds are defined, only edge atoms are
considered 'too near the edge.'
Note that some care must be taken when a building a molecule template
for a given simulation. All atom types in the pre-reacted template
@ -178,23 +202,30 @@ A discussion of correctly handling this is also provided on the
The map file is a text document with the following format:
A map file has a header and a body. The header of map file the
contains one mandatory keyword and one optional keyword. The mandatory
keyword is 'equivalences' and the optional keyword is 'edgeIDs':
contains one mandatory keyword and two optional keywords. The mandatory
keyword is 'equivalences' and the optional keywords are 'edgeIDs' and
'customIDs':
N {equivalences} = # of atoms N in the reaction molecule templates
N {edgeIDs} = # of edge atoms N in the pre-reacted molecule template :pre
N {edgeIDs} = # of edge atoms N in the pre-reacted molecule template
N {customIDs} = # of atoms N that are specified for a custom update :pre
The body of the map file contains two mandatory sections and one
optional section. The first mandatory section begins with the keyword
The body of the map file contains two mandatory sections and two
optional sections. The first mandatory section begins with the keyword
'BondingIDs' and lists the atom IDs of the bonding atom pair in the
pre-reacted molecule template. The second mandatory section begins
with the keyword 'Equivalences' and lists a one-to-one correspondence
between atom IDs of the pre- and post-reacted templates. The first
column is an atom ID of the pre-reacted molecule template, and the
second column is the corresponding atom ID of the post-reacted
molecule template. The optional section begins with the keyword
molecule template. The first optional section begins with the keyword
'EdgeIDs' and lists the atom IDs of edge atoms in the pre-reacted
molecule template.
molecule template. The second optional section begins with the keyword
'Custom Edges' and allows for forcing the update of a specific atom's
atomic charge. The first column is the ID of an atom near the edge of
the pre-reacted molecule template, and the value of the second column
is either 'none' or 'charges.' Further details are provided in the
discussion of the 'update_edges' keyword.
A sample map file is given below:
@ -255,6 +286,18 @@ The {stabilize_steps} keyword allows for the specification of how many
timesteps a reaction site is stabilized before being returned to the
overall system thermostat.
The {update_edges} keyword can increase the number of atoms whose
atomic charges are updated, when the pre-reaction template contains
edge atoms. When the value is set to 'charges,' all atoms' atomic
charges are updated to those specified by the post-reaction template,
including atoms near the edge of reaction templates. When the value is
set to 'custom,' an additional section must be included in the map
file that specifies whether to update charges, on a per-atom basis.
The format of this section is detailed above. Listing a pre-reaction
atom ID with a value of 'charges' will force the update of the atom's
charge, even if it is near a template edge. Atoms not near a template
edge are unaffected by this setting.
In order to produce the most physical behavior, this 'reaction site
equilibration time' should be tuned to be as small as possible while
retaining stability for a given system or reaction step. After a
@ -323,7 +366,7 @@ bond/break"_fix_bond_break.html, "fix bond/swap"_fix_bond_swap.html,
[Default:]
The option defaults are stabilization = no, prob = 1.0, stabilize_steps = 60
The option defaults are stabilization = no, prob = 1.0, stabilize_steps = 60, update_edges = none
:line

6
doc/src/fix_nve_dot.txt
View File

@ -36,8 +36,8 @@ The command is equivalent to the "fix nve"_fix_nve.html.
The particles are always considered to have a finite size.
An example input file can be found in /examples/USER/cgdna/examples/duplex1/.
A technical report with more information on this integrator can be found
"here"_PDF/USER-CGDNA-overview.pdf.
Further details of the implementation and stability of the integrator are contained in "(Henrich)"_#Henrich3.
The preprint version of the article can be found "here"_PDF/USER-CGDNA.pdf.
:line
@ -59,3 +59,5 @@ See the "Build package"_Build_package.html doc page for more info.
[(Davidchack)] R.L Davidchack, T.E. Ouldridge, and M.V. Tretyakov. J. Chem. Phys. 142, 144114 (2015).
:link(Miller1)
[(Miller)] T. F. Miller III, M. Eleftheriou, P. Pattnaik, A. Ndirango, G. J. Martyna, J. Chem. Phys., 116, 8649-8659 (2002).
:link(Henrich3)
[(Henrich)] O. Henrich, Y. A. Gutierrez-Fosado, T. Curk, T. E. Ouldridge, Eur. Phys. J. E 41, 57 (2018).

6
doc/src/fix_nve_dotc_langevin.txt
View File

@ -114,8 +114,8 @@ The scale factor after the {angmom} keyword gives the ratio of the rotational to
the translational friction coefficient.
An example input file can be found in /examples/USER/cgdna/examples/duplex2/.
A technical report with more information on this integrator can be found
"here"_PDF/USER-CGDNA-overview.pdf.
Further details of the implementation and stability of the integrators are contained in "(Henrich)"_#Henrich4.
The preprint version of the article can be found "here"_PDF/USER-CGDNA.pdf.
:line
@ -139,3 +139,5 @@ See the "Build package"_Build_package.html doc page for more info.
[(Miller)] T. F. Miller III, M. Eleftheriou, P. Pattnaik, A. Ndirango, G. J. Martyna, J. Chem. Phys., 116, 8649-8659 (2002).
:link(Dunweg3)
[(Dunweg)] B. Dunweg, W. Paul, Int. J. Mod. Phys. C, 2, 817-27 (1991).
:link(Henrich4)
[(Henrich)] O. Henrich, Y. A. Gutierrez-Fosado, T. Curk, T. E. Ouldridge, Eur. Phys. J. E 41, 57 (2018).

117
doc/src/fix_plumed.txt Normal file
View File

@ -0,0 +1,117 @@
"LAMMPS WWW Site"_lws - "LAMMPS Documentation"_ld - "LAMMPS Commands"_lc :c
:link(lws,http://lammps.sandia.gov)
:link(ld,Manual.html)
:link(lc,Commands_all.html)
:line
fix plumed command :h3
[Syntax:]
fix ID group-ID plumed keyword value ... :pre
ID, group-ID are documented in "fix"_fix.html command :ulb,l
plumed = style name of this fix command :l
keyword = {plumedfile} or {outfile} :l
{plumedfile} arg = name of PLUMED input file to use (default: NULL)
{outfile} arg = name of file on which to write the PLUMED log (default: NULL) :pre
:ule
[Examples:]
fix pl all plumed all plumed plumedfile plumed.dat outfile p.log
[Description:]
This fix instructs LAMMPS to call the PLUMED library, which allows one
to perform various forms of trajectory analysis on the fly and to also
use methods such as umbrella sampling and metadynamics to enhance the
sampling of phase space.
The documentation included here only describes the fix plumed command.
This command is LAMMPS specific whereas most of the functionality
implemented in PLUMED will work with a range of MD codes and also when
PLUMED is used as a stand alone code. The full documentation for PLUMED
is available at "this website"_http://www.plumed.org/documentation
The PLUMED library is developed at
"https://github.com/plumed/plumed2"_https://github.com/plumed/plumed2 A
detailed discussion of the code can be found in "(PLUMED)"_#PLUMED.
There are some example scripts for using this package with LAMMPS in the
examples/USER/plumed directory.
:line
The command to call PLUMED above is reasonably self explanatory. Within
the input file for lammps the user is required to specify the input file
for PLUMED and a file on which to output the PLUMED log. The user must
specify both of these arguments every time PLUMED is to be used.
Furthermore, the fix plumed command should appear in the LAMMPS input
file after the relevant input paramters (e.g. the timestep) have been
set.
The {group-ID} entry is ignored. LAMMPS will always pass all the atoms
to PLUMED and there can only be one instance of the plumed fix at a
time. The plumed fix communicates the minimum amount of information
required and the PLUMED supports multiple, completely independent
collective variables, multiple independent biases and multiple
independent forms of analysis. There is thus really no restriction in
functionality by only allowing only one plumed fix in the LAMMPS input.
The {plumedfile} keyword allows the user to specify the name of the
PLUMED input file. Instructions as to what should be included in a
plumed input file can be found in the "documentation for
PLUMED"_http://www.plumed.org/documentation.
The {outfile} keyword allows the user to specify the name of a file on
which to output the PLUMED log. This log file normally just parots the
information that is contained in the input file. The names of the files
on which the results from the various analyses that have been performed
using PLUMED will be specified by the user in the PLUMED input file.
[Restart, fix_modify, output, run start/stop, minimize info:]
When performing a restart of a calculation that involves PLUMED you must
include a RESTART command in the PLUMED input file as detailed in the
"PLUMED documentation"_http://www.plumed.org/documentation. When the
restart command is found in the PLUMED input PLUMED will append to the
files that were generated in the run that was performed previously.
Furthermore, any history dependent bias potentials that were accumulated
in previous calculations will be read in when the restart command is
included in the PLUMED input.
The "fix_modify"_fix_modify.html {energy} option is not supported by
this fix.
Nothing is computed by this fix that can be accessed by any of the
"output commands"_Howto_output.html within LAMMPS. All the quantities
of interest can be output by commands that are native to PLUMED,
however.
[Restrictions:]
This fix is part of the USER-PLUMED package. It is only enabled if
LAMMPS was built with that package. See the "Build
package"_Build_package.html doc page for more info.
There can only be one plumed fix active at a time. Since the interface
communicates only the minimum amount of information and since the PLUMED
module itself can handle an arbitrary number of analysis and biasing
methods, this is not a limitation of functionality.
[Related commands:]
"fix smd"_fix_smd.html
"fix colvars"_fix_colvars.html
[Default:]
The default options are plumedfile = NULL and outfile = NULL
:line
:link(PLUMED)
[(PLUMED)] G.A. Tribello, M. Bonomi, D. Branduardi, C. Camilloni and G. Bussi, Comp. Phys. Comm 185, 604 (2014)

1
doc/src/fixes.txt
View File

@ -118,6 +118,7 @@ Fixes :h1
fix_phonon
fix_pimd
fix_planeforce
fix_plumed
fix_poems
fix_pour
fix_precession_spin

3
doc/src/group.txt
View File

@ -225,8 +225,7 @@ atomfile-style variable. The variable is evaluated and atoms whose
per-atom values are 0.0, are removed from the dynamic group. If the {property}
keyword is used, the per-atom property name must be a previously defined
per-atom property. The per-atom property is evaluated and atoms whose
values are 0.0 are removed from the dynamic group, otherwise they
are added to the group.
values are 0.0 are removed from the dynamic group.
The assignment of atoms to a dynamic group is done at the beginning of
each run and on every timestep that is a multiple of {N}, which is the

1
doc/src/lammps.book
View File

@ -338,6 +338,7 @@ fix_orient.html
fix_phonon.html
fix_pimd.html
fix_planeforce.html
fix_plumed.html
fix_poems.html
fix_pour.html
fix_precession_spin.html