CUBY logo


Job type

job:    symbol     singlepoint | energy | e | optimize | opt | dynamics | md | charges | interaction | inte | int_e | int_energy | freq | atomization | relative | solvation | dft_sapt | charge_transfer

Job type selection. This keyword accepts several synonyms for each job type:

  • singlepoint (energy, e)
  • optimize (opt)
  • dynamics (md)
  • charges
  • interaction (int_energy, int_e, inte)
  • freq
  • atomization
  • relative
  • solvation
  • dft_sapt
  • charge_transfer

job_name:    string ["Cuby"]

Job name, now used only to name the calculations when submitted to queue


config:    hash

This keyword allows to temporarily modify the config entries.

Input / output

geometry:    string

File with molecular geometry. This is only an optional way how to specify geometry, normaly it is specified from command line.

job_cleanup:    boolean [yes]

Delete directories with files used for the calculation when cuby successfuly ends.

verbosity:    integer [3]

print:    string [""]

List of extra information to be printed, depending on context and type of calculation.

  • Fragmentation - fragment energies and interactions: "frag[ment] e[nergies]"
  • Restraints modifier - energy of the restraints: restr[aints]
  • Optimization - print cartesian gradient in each step: "gradient"
  • Frequencies - list complex eigenvalues in addition to frequencies: "eigen[values]"

energy_decomposition:    boolean [no]

Print available energy components in addition to the total energy

energy_component:    string

Replace the total energy in output (and any result of energy calculation) by energy component of a given name. The original energy is saved in the energy components and can be displayed by energy_decomposition keyword.

energy_component_add:    string

Add or subtract energy components of a given names to/from the total energy (any result of energy calculation). The original energy is saved in the energy components and can be displayed by energy_decomposition keyword. The value of this keyword should be a comma-delimited list of energy component names (as seen in energy decomposition). Components to be subtracted instead of added should have prefix "-")

results_yaml:    boolean [no]

Write important output to a yaml file. Works for following modules:

  • cuby_freq - writes freq.yaml

Calculation setup

use_old_results:    boolean [no]

Use old results in the calulation directory if they exist instead of calculating them again. Works with interfaces:

  • Turbomole
  • DFTB

mem:    integer [400]

Memory limit (in MB) passed to the calculations.

charge:    integer [0]

Charge of the system

multiplicity:    integer [1]

Multiplicity of the system

interface:    symbol

Interface used for calculation. For possible values, see the page on interfaces.

modifiers:    string [""]

Modifiers of potential (calculations added to the one specified in interface). Some interfaces can be used as modifiers. Entered as comma separated list of interface names.

point_charges:    string [""]

External point charges file for calculations in electrostatic field defined by point charges.

atomic_charges:    boolean [no]

Atomic charges calculation - request calculation of atomic charges in addition to energy calculation.

atomic_charges_write:    string

File to write atomic charges to, no output to file when not specified (just to std. output).

start_from_previous:    boolean [true]

Options that allows to disable starting the calculation from previous step of optimization / MD. Currently available for interfaces:

  • dftb
  • metainterface: PM6

parallel:    integer [1]

Reques parallel run of external code on specified number of processors. Available for interfaces:

  • dftb
  • turbomole
  • molpro (multithreading only, not MPI)

parallel_mode:    symbol [none]     mpi | shared_memory

Turbomole only. Switch between MPI and shared memory parallelization.

parallel_threads:    integer [1]

parallel_gpu:    integer [1]

Number of GPUs to use.

solvent_model:    symbol [none]     none | cosmo | gbm | igb7

Implicit solvent model:

  • none
  • COSMO (in Turbomole & Molpro) - COnductor-like Screening MOdel
  • GBM (in AMBER) - Generalized Born Model
  • IGB7 (in AMBER) - Another Generalized Born

solvent_epsilon:    float [78.5]

Dielectric constant of implicit solvent

numerical_hessian:    boolean [no]

Numerical hessian calculation instead of analytical one

numerical_hessian_d:    float [0.001]

Atom displacement used in numerical calculation of hessian (Angstrom)

numerical_hessian_oneside:    boolean [no]

Calculate the hessian from only one displacement per coordinate per atom.

partial_hessian:    string

Selection of atoms for which numerical hessian is calculated.

partial_hessian_full_geometry:    boolean [yes]

Write full geometry to molden.input (otherwise, only nonfrozen part is written)

script_after_calculation:    string [""]

Run a script after calculation is finished (provide valid name of an executable file)

Distributed calculations

distributed:    boolean [no]

Parallelization by the means of running multiple partial calculation on more processors, using an internal queue system.

distributed_remote:    boolean [no]

If enabled, distributed calculation uses multiple nodes in network. Details on communication with the node must be specified by additional keywords.

distributed_protocol:    symbol [ssh]     ssh | ssh_sharedfiles

Specifies protocol used to access other nodes for distributed calculations.

  • ssh - Use ssh to login to nodes, use scp to transfer files (paswsordless authentication must be set up)
  • ssh_sharedfiles - The same as ssh, without copying the files. Use where the calculation directory is shared over the network.

distributed_calcdir:    string

Temporary (scratch) directory on nodes, to where the files are calculated for remote distributed calculation.

nodes:    string

List of nodes vavailable for remote distributed calculation. Possible input is:

  • "(list) node_name1, node_name2, ..." - direct specification of node names
  • "(filename) name_of_file" - read list of nodes at the time of the calculation (text file, one node name per line)

The evaluation of the filename uses full shell expansion - it is possible to read the filename from environment variable provided by the cluster - e.g. "(filename) $PBS_NODEFILE" on clusters using PBS queue system.

nproc:    integer [1]

Number of processors per node (applies both to local and remote calculations)

Interfaces: common keywords for qm methods

method:    symbol     hf | dft | mp2 | lmp2 | mp2half | mp2c | mp3 | ccsd | ccsdt | ccsd_t | "ccsd(tq)" | ccsd_tq | ccsd_t_q | ccsd_t_qp | ccsd_t_q_p | fci | mp2_f12 | ccsd_f12 | ccsdt_f12 | am1 | pm3 | pm6 | pm7 | rm1 | custom

Selection of the calculation method
Used by:

  • Molpro
  • Mopac
  • PQS

basisset:    string

Basisset for the calculation
Used by:

  • Demon
  • Molpro
  • PQS
  • Turbomole

basisset_elements:    hash

Separate specification of basis set for different elemets
Used by:

  • Demon
  • Molpro
  • Turbomole

auxiliary_basis:    string

Auxiliary basis set name. When not present, default is used. In Turbomole, this keyword can be used only when default auxiliary basis for given basis set is not defined in the library.
Used by:

  • Demon
  • Molpro
  • Turbomole

auxiliary_basis_elements:    hash

Separate specification of basis set for different elemets
Used by:

  • Molpro (MP2 only)

functional:    string

Functional for DFT calculations
Used by:

  • Demon
  • Fhi_aims
  • Molpro
  • PQS
  • Turbomole

scf_integrals:    symbol [conventional]     conventional | direct | multipole

Integral calculation / storage mode.
Used by:

  • Demon
  • Molpro (only conventional / direct)
  • NWChem (only conventional / direct)

scf_cycles:    integer [60]

Max. number of SCF cycles
Used by:

  • Demon
  • Fhi_aims
  • Turbomole

scf_limit:    integer [6]

Convergence limit for SCF calculation, in form 1e-X Hartree
Used by:

  • Demon
  • Molpro
  • PQS
  • Turbomole

levelshift:    float

Fixed levelshift, in Hartrees. No fixed levelshift if not used.
Used by:

  • Demon
  • Turbomole

atomic_charges_method:    symbol     mulliken | nbo | loewdin | becke | hirshfeld_vdw | hirshfeld_covalent | hirshfeld_density

Type of atomic charges to be calculated

  • mulliken: turbomole, deMon
  • nbo: turbomole
  • loewdin: deMon
  • becke: deMon
  • hirshfeld_*: deMon

Interfaces: common keywords for other methods

polarized:    boolean [no]

Polarized calculation of fragments/QM part, uses point charges from the rest ot the system
Used by:

  • QMMM
  • fragmented
  • mfba

atomic_charges_read:    string

Name of file containing atomic charges of use with polarized keyword.
Used by:

  • QMMM
  • fragmented
  • mfba



metainterface:    symbol

Name of metainterface definition (read from data/metainterfaces.yaml)

mtif_keywords:    string [""]

Extra keywords that are directly included in the input file built by metainterface. CAn be used in these metainterfaces:

  • PM6

mtif_extras:    string [""]


dispersion_para:    string

Dispersion parameters (comma seperated numbers, third paremater is optional): radii scaling factor (sr), exponent (alpha) and global scaling (s6). When not specified, cuby attempts to use default values for the given method (defined in file data/dispersion.yaml).

dispersion_elements_r:    hash

Override element radii from teh database with custom values (in Angstrom)

dispersion_elements_c:    hash

Override c6 coefficients (unit: same as in dispersion.yaml)

dispersion_mix:    symbol [jurecka]     jurecka | grimme

Mixing of vdW eadii and C6 cooefficient in dispersion calculation. The default parameters must be used with Jurecka's mixing.

dispersion_db_file:    string

Custom file with dispersion parameters to be used instead the default file data/dispersion.yaml

dispersion_hyb: [no]    boolean

Use hybridization-dependent parameters


h_bonds_damped:    boolean [no]

Switches on damping of the H-bond correction (inflex point of damping function 2.5 A). This is just an experimental option, the correction was parameterized without the damping.

h_bonds_new:    boolean [no]

Use newer vesrion of the correction (negligible improvement but better code)

H-bonds v.2.0

h_bonds2_parameters:    symbol [from_method]

Name of the parameter set. Available sets are:

  • pm6
  • om3
  • am1
  • dftb

h_bonds2_parameter_file:    string

File with parameters for H0bonding correction. When not provided, default file contained in cuby is used.

h_bonds2_fixed_charges:    boolean [no]

Use charges from the first cycle for the remainder of the optimization/MD

h_bonds2_atomtype_charges:    hash

Define charges for atom types.

h_bonds2_search_cutoff:    boolean [no]

Use distance cutoff 7 Angstrom when building the list of possible H-bonds

h_bonds2_calc_cutoff:    boolean [no]

Smooth cutoff function switching the correction to zero between h_bonds2_calc_cutoff_r and h_bonds2_calc_cutoff_r + 1 A.

h_bonds2_calc_cutoff_r:    float [6.0]

Cutoff distance in h_bonds2_calc_cutoff:

H-bonds v.4.0

h_bonds4_parameters:    hash

Parameters for the H-bond correction. By default, parameters for the selected method are loaded automatically.

h_bonds4_scale_charged:    boolean [yes]

Scaling of the correction for charged groups (COO-, NH3+).

h_bonds4_extra_scaling:    hash

Manual definition of additional chrged sites to be scaled. Key of the hash is selection of donor/acceptor atoms to be applied to, value is a sacling factor.

h_bonds4_pt_corr:    float [0.0]

Increase of proton transfer barriers

h_bonds4_skip_acceptor:    array [[]]

Skip corerction for specific acceptor ararngement. Following cases are supported:

  • OS - oxygen bound to sulfur


amber_gbm_saltcon:    float [0.0]

Concentration of virtual salt on GBM solvent

leaprc:    string

Leaprc prividing the forcefield for preparation of AMBER calculation. When not specified, the value from configuration file is used. More leaprc files can be provided using a multiline string or list separated by ':' or ';', they are sourced one after another in tleap when the calculation is being prepared. In the leaprc files, string '%cuby' is substituted with cuby installation directory and %amberhome is substituted with environment variable AMBERHOME. Cuby looks for leaprc in these places:

  • filesystem root directory (it means absolute path should be provided)
  • $cuby_install_dir/data/qmmm
  • $AMBERHOME/dat/leap/cmd

amber_input:    string

Custom specification of AMBER input file, by default it is created by the interface.

amber_old_top:    boolean [no]

Write AMBER topology in old format

amber_gromacs:    boolean [no]

Use GROMACS to perform the calculation (using ambconv)


tm_level:    symbol     hf | rihf | rijkhf | dft | ridft | mp2 | rimp2 | ricc2_scs_mi_mp2 | ricc2_cc2 | ricc2_b2plyp | ricc2_ccsdt | ricc2_rimp2

Level of calculation.

scf_denconv:    float [1.0e-6]

Convergence limit for density in SCF calculation

tm_basis_file:    string

Supply a basis not known by turbomole as a file

dft_grid:    string ["m3"]

Grid for DFT calculations

tm_marij:    boolean [no]

Multipole accelerated RI-J for RI-DFT

tm_delmos:    boolean [no]

Delete orbital file after calculation

tm_b2plyp_disp:    boolean [yes]

auxiliary_basis_jk:    string

Auxiliary basis set name for RI-JK in Turbomole. When not present, default is used.


dftb_dispersion:    boolean [yes]

The original dispersion correction as implemented in the dftb+ code

dftb_3rd_order:    boolean [no]

Third order expansion of DFTB energy. Requires DVTB+ ver 1.1

dftb_hubbard_derivs:    hash [{H:]

3rd order DFTB uses derivatives of Hubbard parameter. The default values are those obtained from DFT calculation

dftb_xh_damping:    boolean [no]

XH electrostatic interaction damping to improve hydrogen bonds. Requires DVTB+ ver 1.1

dftb_xh_damping_exp:    float [4.95]

parameter (exponent) for dftb_xh_damping . Default value is taken from paper on the method optimized for use with MIO set.

dftb_scc_convergence:    float [1.0e-05]

Self-consistent charges calculation convergence limit

dftb_slko_path:    string

Path to Slater-Kostner files, default is read from config file

dftb_e_temp:    float [0.0]

Electronic temerature

dftb_blur_charges:    float [0.0]

Half-width of gaussian blur of external point charges

dftb_safe:    boolean [no]

Use more robust eigensolver


fragments:    array

Specification of fragments. The format is YAML array of atom selections

fragment_charges:    array

Specification of charges of the fragments. Format: YAML array of integers. If ommited, all charges are set to zero.

fragmentation_ch_mindist:    float [0.0]

Remove point charges closer to atoms than this distance

fragmentation_3rd_order:    boolean [no]

Third-order calculation using all trimers

fragmentation_fullbasis:    boolean [no]

All fragments are calculated in basis set of the whole system

fragmentation_corr:    boolean [no]

Link-link interaction correction (electrostatics from point charges)

write_interaction_matrix:    string

File to write list of pairwise interactions between fragments, that can be visualized by script visualize_intmatrix.

fragmentation_link_dist:    float [1.09]

Bond distance for construction the hydrogen caps.

fragmentation_fixed_links:    boolean [no]

Bond distance to link can be fixed rather than to be derived from length of the original bond

fragmentation_fix_grad:    boolean [yes]

Experimental: Fix gradients in dimers


mfba_levels:    integer

Number of levels used for multiscale calculation. Minimum is 2. Each level requires specification of distance threshold and method of calculation in its own subsection of the input. The subsections are named:

  • mfba_method_hi
  • mfba_method_lo_1
  • mfba_method_lo_2
  • ...

mfba_threshold:    float [999999.9]

Thershold is maximum distance between fragments at which current method is applied. In addition to this specification, all covalent-bound pairs are calculated at the high level.


qmmm_core:    string

Selection of QM core in QM/MM calculation

qmmm_cut_bonds:    array [[]]

List of bonds that are cut and replaced by link atoms

qmmm_qm_charge:    integer [0]

Charge of QM region within QM/MM calculation

qmmm_geometry_only:    boolean [no]

If set, cuby will print geometry of the QM cluster to file "qmmm_cluster.pdb" and exit. Use this feature to prepare MM parameters for the cluster.

qmmm_rename_residues:    array [[]]

List of residues to be renamed in QM region. Format: YAML array of strings "selection new_name".

qmmm_add_ter:    string [""]

Selection of atoms in QM regieon (selection expression) after which PDB TER labels are added.

microiterations:    symbol [never]     each | first | never

Optimize the MM region (QM part frozen) before each QMMM step. This should be combined with optimization of the QM region and cut bonds (using optimize_region keyword).

microiterations_maxcycles:    integer [10000]

Max. number of steps for microiterations (microiterations should converge in each QM/MM step).

qmmm_remove_charges:    string [""]

Selection of MM atoms not to be used as point charges in polarized calculation. By default, charges are removed from the atoms in the bonds across QM/MM boundary, this option can only add to this.


multi_qmmm_n:    integer

Number of QM regions


qmmm_xs_center:    string

Selection of atoms defining the center of the solvent sphere

qmmm_xs_radius:    float

Radius of the solvent sphere

qmmm_xs_buffer:    float [0.2]

Thickness of the buffer layer where water molecules can exchange

qmmm_xs_water_energy_qm:    float [0.0]

Energy of a single water molecule in given method. Used to correct the difference between QM and MM

qmmm_xs_water_energy_mm:    float [0.0]

Energy of a single water molecule in given method. Used to correct the difference between QM and MM


waterball_type:    symbol     harmonic | beglov_roux

Waterball boundary potential type

waterball_radius:    float

Radius above which harmonic potential is applied (only for watreball_type: harmonic)

waterball_fconst:    float [0.001]

Force constant for watreball_type: harmonic

waterball_center:    symbol     origin | solute | fixed_solute

Center of the sphere:

  • Origin - Origin of coordinate system
  • Solute - Center of mass of solute, specified by waterball_solute
  • Fixed_solute - Origin of coordinate system, solute held there by additional harmonic potential

waterball_solute:    string

Selection defining the solute


gaussian_keywords:    string

Gaussian job specification (jsut the method, without # and what to calculate)

gaussian_extras:    string [""]

Extra gaussian input (added after the geometry and blank line)

gaussian_overlay:    string [""]

Extra gaussian input in the overlay section


demon_basis_file:    string

File containing definition of basis sets. If not specified, default file (file BASIS from directory specified in config) is used. The filename is firstly used as is, and when no file is found, the demon basis directory from the config file is prepended

demon_auxis_file:    string

File containing definition of auxiliary basis sets. If not specified, default file (file AUXIS from directory specified in config) is used.

demon_keywords:    string [""]

Extra keywords for demon (single or multiline string)

demon_cdf_limit:    integer [4]

Additional SCF converge limit: error of coefficients (1.0e^-value)

demon_grid:    string ["MEDIUM"]

Grid selection. Allowed values are FINE|MEDIUM|COARSE or a number that define the tolerance (integer X, used as 1.0e-X a.u.)

constrained_dft:    boolean [no]

cdft_population:    symbol [becke]     mulliken | loewdin | becke | hirshfeld | van_alsenoy | vdd

cdft_tolerance:    integer [8]

cdft_force_converged_constraints:    boolean [no]

cdft_maxcycles:    integer [50]

cdft_hda:    boolean [no]

cdft_radii_gauss:    float [1.0]

cdft_radii:    symbol [gaussian]     gaussian | atomic_density | fragment_density

cdft_constraints:    array

cdft_multipliers_guess:    symbol [none]     none | previous | extrapolate

cdft_auxis_grad:    boolean [no]

Use auxiliary density for CDFT gradients

pseudopotentials:    string

Pseudopotential specification

pseudopotentials_elements:    hash

Separate specification of pseudopotentials set for different elemets

demon_ecps_file:    string

File containing definition of pseudopotentials. If not specified, default file (file ECPS from directory specified in config) is used.

demon_mcps_file:    string

File containing definition of pseudopotentials. If not specified, default file (file MCPS from directory specified in config) is used.

demon_2nd_state_options:    string

Options (in demon format) for the CUBY keyword to request simulatenous calculation of a second state.


restraints:    array

List of restraints. Each line has format: distance|angle|torsion; value; force_constant; atom_selection1; atom_selection2; ... Angles are specified in degrees

Cartesian restraints

cartesian_restraints:    array

List of restraints. Each entry is a hash:

  • selection: atom selection expression
  • center: centertype (allowed values: 'initial', 'origin' or 'coord(x,y,z)')
  • r_plateau: radius of flat region around center (default value 0.0)
  • fconst: harmonic force constant, in kcal/mol/A (default value 10.0)

Simple restraints (cartesian)

simple_restraints_sel:    string

Selection of atoms to be restrained to their initial coordinates

simple_restraints_f:    float

Force constant (kcal/mol/angstrom^2)

simple_restraints_plateau:    float [0.0]

Plateau radius


fhi_aims_level:    symbol [tight]     light | tight | really_tight

Level of accuracy of a FHI-aims calculation, contating both basis set and grid setup.

fhi_aims_dispersion:    boolean [yes]

Use the dispersion correction using the on-the-fly derived c6 coefficients

fhi_aims_read_hirshfeld:    boolean [no]

Read results of the Hirshfeld analysis

Parallel run using mpi (specify a script that runs the mpi job in config)

fhi_aims_scalapack:    boolean [no]

Potential mixer

mixer_ratio_a_to_b:    float

Ratio for mixing the two calculations A and B. Use value between 0.0 and 1.0

mixer_transmutations:    hash

List of atoms for which element is changed. Use a yaml hash representation where the key is the seection expression and value is the new element.


electronic_spin:    symbol     restricted | unrestricted [restricted]

density_fitting:    boolean [no]

Density fitting for DFT and MP2 calculations

direct_corr:    boolean [no]

Direct calculation of correlation energy. HF is controlled separately by keyword scf_integrals. Available for MP2 and CCSD only.

molpro_basis_file:    string

User-specified basis from external file (used instead of basis keyword)

dftsapt_frozen_core:    boolean [no]

Use frozen core option, number of frozen shells is determined automagically

dftsapt_shift_func:    string

Functional used to calculated the shift for monomers. If not specified, the same functional as for SAPT is used.

dftsapt_steps:    symbol     all | shifts | sapt [all]

Stepts of dft-sapt aclculation:

  • all - full calculation
  • shifts - calculate only shifts and stop
  • sapt - skip shift calculations (read values from the input), run SAPT

dftsapt_shift_a:    float

Shifts for DFT-SAPT calculations for molecule A (kcal/mol). When dftsapt_steps == all (default), shifts are calculated automatically.

dftsapt_shift_b:    float

Shifts for DFT-SAPT calculations for molecule B (kcal/mol). When dftsapt_steps == all (default), shifts are calculated automatically.

dftsapt_shift_basisset:    string

Basis set for the calculation of the shift. If not set, the same basis as for SAPT calculation is used.

sapt_level:    integer [3]

f12_ansatz:    string ["3C(FIX)"]

Explicitly correlated wave function ansatz. See molpro manual for possible values.

f12_ccsd:    symbol [a]     a | b

Approximation type in CCSD-F12 calculation. Type A is recommended for smaller basis set (DZ and TZ), B for larger ones.

f12_geminal_exp:    float [1.0]

molpro_frozen_core:    boolean [yes]


forcefield_file:    string

YAML file containing the definition of the forcefield

Composite calculations

composite_method_names:    array

List of names of the elementary calculations. The name could contain only lowercase letters, numbers and the undescore character, and it must begin with a letter. These name are used in two other places in the input:

  • To name subsection defining the methods i.e. the subsection name for method xyz is calculation_xyz
  • As variable names in the equation (keyword composite_equation)

composite_equation:    string

Equation used to calculate the final energy and gradient.

Dft-d3 using grimme's code

dftd3_old:    boolean [no]

Use DFT-D2 functions and parameters


disp3_damped:    boolean [yes]

This keyword can be used to switch off the damping function

disp3_damping_bj:    boolean [no]

Becke-Johnson damping function

disp3_fixhyb:    boolean [no]

Use fixed integer hybridization instead of continuous one proposed by Grimme

disp3_3body:    boolean [no]

Add 3-body term

disp3_cutoff:    float

When set, distance cutoff with specified radius is applied.

disp3_s6:    float

Scaling of yhe C6 term

disp3_s8:    float

Scaling of yhe C8 term

disp3_sr6:    float

Radii scaling for damping function in C6 term

disp3_sr8:    float

Radii scaling for damping function in C8 term

disp3_alpha6:    float

Exponent in damping function in C6 term

disp3_alpha8:    float

Exponent in damping function in C8 term

disp3_bj_a1:    float

Parameter a1 for Becke-Johnson damping function

disp3_bj_a2:    float

Parameter a2 for Becke-Johnson damping function

disp3_hh_fix:    boolean [no]

H-H repulsion for semiempirical methods

disp3_hh_fix_version:    integer [1]

Value 2 enables the new HH repulsion.

disp3_hh_para:    hash

H-H repulsion parameters


mopac_mozyme:    boolean [no]

MOZYME orbital localization

mopac_d_corr:    boolean [no]

Dispersion correction calculated by MOPAC

mopac_h_corr:    boolean [no]

H-bond correction calculated by MOPAC

mopac_dh_plus:    boolean [no]

PM6-DH+ correction by Martin Korth

mopac_custom_method:    string

Method specification in MOPAC foprmat used when method is set to 'custom'

H-bond correction calculated by MOPAC

mopac_keywords:    string [""]

Extra keywords added to the input built by cuby

mopac_setpi:    array [[]]

Manual specification of double bonds for MOZYME, array of atom pairs (atom numbers separated by semicolon)

mopac_setcharge:    hash [{}]

Manual specification atomic charges for MOZYME, hash of charges indexed by atom number

mopac_peptide_bond_fix:    boolean [no]

Controls the corerction for peptide bond torsion. It was introduced as an addition to the RM1, PM6 and PM7 methods, therefore it is switched off by default for backwards compatibility. MOPAC documentation:

mopac_parallel:    boolean [yes]

The latest version allows parralelization, but the input can not be used with older versions of MOPAC2012. Disable this option when older version is used.


charmm_psf_file:    string

User must provide PSF file prepared by charmm. The geometry used for the calculation must be identical to the one described by the PSF file.

charmm_ff_top:    string

Forcefield definition file: topology

charmm_ff_par:    string

Forcefield definition file: parameters

charmm_extras:    string [""]

Extra options for charmm, included directly into the output before the actual calculation.

charmm_stream_files:    array [[]]

Load stream file(s) with parameters.

Empirical bsse correction

bsse_corr_parameters:    hash

Friesner dft correction

friesner_parameter_set:    string

Following parameter sets are available:

  • LACVP*
  • aug-cc-pVDZ
  • aug-cc-pVDZ-CP

friesner_parameter_file:    string

Name of yaml file with parameters

X-bonds dftb

dftb_xbond_parameters:    hash [{]

Parameters for the X-bond correction. By default, parameters for the selected method are loaded automatically.


use_symmetry:    boolean [no]

Enable symmetry in calculation


aces_basis_file:    string

Use custom GENBAS file. If not specified, default file specified in the config is used.

Supply a basis not known by turbomole as a file

aces_cc_safe:    boolean [no]

Use more robust CC iteration accelerator

aces_delete_big_files:    boolean [yes]

Delete large integral files after a successful calculation to save disk space

aces_keywords:    string [""]

Extra ACES keywords to be added into the input

Ljdisp + bsse correction for dft

ljdisp_epsilon:    hash

ljdisp_sr:    float

ljdisp_sr0:    float [1.0]

ljdisp_repulsion:    array [[]]

Each item should be an array of: element1, element2, energy, radius, width


lookup_table:    string ["lookup_table"]

Name of the yaml file containing the lookup table data. Current directory and cuby_dir/data/lookups are searched in this order.

Gcp bsse correction

gcp_parameter_file:    string

Parameter file for the gcp program. When the keyword is not set, default paramteres for the method and basis will be used if available.

Geometry optimization

coordinates:    symbol [cartesian]     cartesian | internal

Coordinates for geometry optimization

maxcycles:    integer [200]

Maximum number of cycles of optimization or molecular dynamics

algorithm:    symbol [rfo]     sd | cg | quasi_newton | rfo | trim | fire | lbfgs

Optimization algorithm

  • SD: Steepest descent
  • CG: Conjugated gradients
  • QUASI_NEWTON: Plain quasi-Newton algorithm (faster, but less efficient than RFO and TRIM)
  • RFO: Rational function to get step, quasi-Newton hessian update
  • TRIM: Optimization algorithm with exact trust radius solver (either Newton or quasi-Newton), allows to optimize TS using image function approach
  • FIRE: (Fast Inertial Relaxation Engine) - MD-like optimizer, Phys.Rev.Lett. 97, 170201 (2006)
  • LBFGS: Limited-memory BFGS

lbfgs_n:    integer [30]

Length of history in LBFGS algorithm. The original default value of 3 is not optimal (larger should be used), and was changed to 30.

opt_use_hessian:    symbol [never]     each | first | never

For TRIM driver only

  • EACH: Newton optimizer, calculate hessian in each step
  • FIRST: Calculate hessian once, then use BFGS update
  • NEVER: Use guess for the initial hessian, then use BFGS update

opt_update_formula:    symbol [dfp_bfgs]     bfgs | dfp | dfp_bfgs | broyden | sr1

Approximate hessian update formula. Works for QUASI_NEWTON, RFO AND TRIM (in QN mode) optimizers:

  • DFP_BFGS - robust combination of DFP and BFGS hessian update
  • The BGFS formula only

Other options are experimental and not intended for routine use

Optimize to transition state (saddlepoint). Works only with TRIM algorithm.

optimize_ts:    boolean [no]

Optimize to transition state (saddlepoint). Works only with TRIM algorithm.

follow_eigenvector:    string ["lowest"]

Eigenvector to follow it TS optimization (TRIM algorithm). Allowed values:

  • "lowest": Allways follow the most negative eigenvector (or a lowest positive)
  • "highest_negative": Allways follow the least negative eigenvector (or a lowest positive)
  • "read": Read in a vector using keyword optimize_ts_evec_guess
  • number (0 and up): In first step, select (lowest - N)th eigenvector, in following steps, follow the one with best overlap with previous.

follow_eigenvector_update:    boolean [no]

Update the eigenvector to follow. If on, new vector is basen on match with the vector used in previous step, otherwise it is selected by comparison with the initial guess.

optimize_ts_evec_guess:    string

Guess of the TS eigenvector with negative curvature (.xyz file containing two or three geometries)

numerical_gradient:    boolean [no]

Numerical gradient calculation instead of analytical one

numerical_gradient_step:    float [0.01]

Displacement used for numerical gradient calculation (Angstrom)

trust_radius:    float [0.1]

Trust radius in Angstrom for QUASI_NEWTON, RFO and TRIM optimizer

opt_tr_update:    boolean [yes]

Update the trust radius dynamically (it is lowered when bad steps are taken, but it can't exceed the value trust_radius_max). Works for QUASI_NEWTON, RFO AND TRIM optimizers.

trust_radius_max:    float [0.3]

Upper limit of trust_radius when dynamic change of it is allowed (by opt_tr_update)

trust_radius_min:    float [1.0e-7]

Lower limit of trust_radius when dynamic change of it is allowed (by opt_tr_update)

opt_refuse_steps:    boolean [yes]

Discard wrog steps and try new step from previous geometry (applies only when opt_tr_update is on)

hessian_estimate:    symbol [diagonal_cart]     diagonal_cart | diagonal_internal | fischer_almlof | lindh | swart | numerical

Initial hessian for optimization:

  • diagonal_cart - Diagonal in cartesian coordinates (= not very good) (default for optimizations of large molecules in cartesian coordinates)
  • diagonal_internal - Diagonal in internal coordinates (default for optimizations in internal coordinates, can be used in cartesians , but the transformation is slow for large molecules)
  • lindh - more advanced version of diagonal_internal (default for optimizations of small molecules in cartesian coordinates when binary extensions are present)
  • fischer_almlof - another version of diagonal_internal
  • swart - another version of diagonal_internal
  • numerical - hessian calculation using method specified in subsection initial_hessian_calculation

hessian_from_file:    string

Load initial hessian from file

hessian_force_positive:    boolean [yes]

Makes the initial hessian positive by setting the elements in diagonalized hessian to their absolute value. This was found to be the most efficient way to fix hessians calculated

File to write optimization history or MD trajectory

history_file:    string [""]

File to write optimization history or MD trajectory

history_freq:    integer [1]

Period of history recording (each Nth cycle). Set to 0 fro no history recording.

history_selection:    string ["all"]

Only the selected atoms from the geometry are written to the history file.

restart_file:    string

Restart file - geometry at the end of calculation. File type is determined from thi file suffix:

  • .zmat - Z-matrix will be written if the geometry was read in z-matrix format, otherwise XYZ format is used
  • .pdb - PDB format
  • any other file name: XYZ format

opt_quality:    float [1.0]

One-number control of optimization quality - multiplier of convergence limits.

opt_convlimit_e:    float [0.006]

Convergence limit for energy change between steps (kcal/mol). If set to 0, this criterium is not evaluated.

opt_convlimit_max_g:    float [1.2]

Convergence limit for maximum gradient component (kcal/mol/A). If set to 0, this criterium is not evaluated.

opt_convlimit_abs_g:    float [0.0]

Convergence limit for abs. value of gradient (gradient norm) (kcal/mol/A). If set to 0, this criterium is not evaluated.

opt_convlimit_rms_g:    float [0.6]

Convergence limit for root mean square of gradient (gradient norm) (kcal/mol/A). If set to 0, this criterium is not evaluated.

opt_convlimit_max_dx:    float [0.0]

Convergence limit for maximum step component (A). If set to 0, this criterium is not evaluated.

opt_convlimit_abs_dx:    float [0.0]

Convergence limit for abs. value of step (A). If set to 0, this criterium is not evaluated.

opt_convlimit_rms_dx:    float [0.0]

Convergence limit for root mean square of step size (A). If set to 0, this criterium is not evaluated.

opt_criteria_baker:    boolean [false]

freeze_atoms:    string [""]

Selection of atoms to be frozen by setting gradients to 0.

freeze_x:    string [""]

Selection of atoms to be frozen by setting gradients in x axis to 0.

freeze_y:    string [""]

Selection of atoms to be frozen by setting gradients in y axis to 0.

freeze_z:    string [""]

Selection of atoms to be frozen by setting gradients in z axis to 0.

remove_translation:    boolean [yes]

Removes translational component of a gradient in each cycle of optimization or a velocity in MD simulation.

remove_rotation:    boolean [yes]

Removes rotational component of a gradient in each cycle of optimization or MD a velocity in MD simulation.

optimize_region:    string

Hard freeze - do not propagate selected atoms other than the selection

Internal coordinates

extra_redundant:    float [0.0]

Adds extra distance coordinates (bonds) between all atoms within specified range

delocalized_internal_coord:    boolean [no]

Use delocalized internal coordinates instead of redundant internal coordinates

Interaction energy calculation

bsse_correction:    boolean [no]

Counterpoise correction in calculation of interaction energy. Only some interfaces support this option.

interaction_decomposition:    boolean [no]

Print energies of the subsystems

selection:    string [auto]

Subsections molecule_a and molecule_b are used to define subsystems. Each can contain :charge and :selection keywords

  charge: 0
  selection: 1-5
  miltiplicity: 2
  charge: -1
  selection: 6-10

Molecular dynamics

timestep:    float [0.001]

Timestep for MD (ps)

velocities:    symbol [random]     random | zero | read

random_seed:    integer

Random seed used to initialize the random number generator when setting up a MD run. When not set, random initialization is used.

velocities_file:    string

history_extended:    symbol [no]     no | xyz_velo

Extended format of trajectory file (use xyz_velo to include velocities)

init_temp:    float [10.0]

Temperature used to generate initial random velocities

thermostat:    symbol [none]     none | berendsen | andersen | nose_hoover | bussi

Selection of thermostat algorithm

temperature:    float [300.0]

Temperature to be maintained by the thermostat

thermostat_tc:    float [0.5]

Thermostat time constant (ps) setting strength of the coupling to the thermostat. The exact mening of the value is different in different algorithms.

temperature_profile:    hash

freeze_hard:    string

Nudged elastic band

neb_images:    integer [20]

Number of images along the minimum energy path

neb_fconst:    float [20.0]

Force constant of spring connecting the images

neb_mep_geometries:    string [""]

Name of file to write the final minimum energy path geometries to.

Harmonic frequencies: cuby_freq

write_hessian:    string

Name of file to write hessian to.

freq_write_molden:    boolean [yes]

Write file molden.input for visualization of normal modes in molden

atom_mass:    hash

Allows to manually assign atom masses. Hash, index is selection of atoms, value is new mass (g/mol).

freq_low_modes_as_rotations:    boolean [no]

Low mode entropy correction by Grimme

Scans: cuby_scan

scan_type:    symbol     zmatrix_setvar | zmatrix_interpolate | cartesian_interpolate

zmat_var_names:    array

zmat_var_values:    array

scan_steps:    integer

Number of steps of coordinate interpolation

scan_reuse_calc_dir:    boolean [no]

Conformation energy

reference_geometry:    string

Reference geometry

Amber resp script

antechamber_names:    symbol     gaff | amber

resp_method:    symbol [hf]     hf | dft | dft_pcm

Development, temporary parameters

development:    hash [{]

A container for temporary parameters to be used in development