oxDNA README ------------ oxDNA is a simulation code that was initially conceived as an implementation of the coarse-grained DNA model introduced by T. E. Ouldridge, J. P. K. Doye and A. A. Louis (http://dx.doi.org/10.1063/1.3552946). It has been since reworked and it is now an extensible simulation+analysis framework. It natively supports DNA, RNA, Lennard-Jones and patchy particle simulations on both CPU and NVIDIA GPUs. REQUIREMENTS ------------ Without CUDA support: cmake >= 2.6 make gcc >= 4.2 / icc >= 11.0 With CUDA support: cmake >= 2.8 make gcc >= 4.2 / icc >= 11.0 CUDA toolkit >= 4.0 Generating the documentation (with 'make docs', see below) requires doxygen. COMPILING oxDNA --------------- Extract the oxDNA archive and then: cd oxDNA # enter the oxDNA folder mkdir build # create a new build folder. It is good practice to compile out-of-source cd build cmake .. # here you can specify additional options, see next section make -j4 # compile oxDNA. The -jX make option makes it compile the code in parallel by using X threads. At the end of the compilation three executables (oxDNA, DNAnalysis and confGenerator) will be placed in the build/bin folder. CMAKE OPTIONS ------------- -DCUDA=ON Enables CUDA support -DDebug=ON Compiles with debug symbols and without optimisation flags -DG=ON Compiles with debug symbols + optimisation flags -DINTEL=ON Uses INTEL's compiler suite -DMPI=ON Compiles oxDNA with MPI support MAKE TARGETS ------------- make Compiles oxDNA make docs Produces html doxygen documentation. oxDNA: DOCS/html_oxDNA/index.html UTILS: DOCS/html_UTILS/index.html TESTING ------- make test Runs quick tests to check whether oxDNA has been correctly compiled or not. This command runs a several short system test in the TEST/ directory Test results can be viewed in TEST/logs make test_scientific Runs longer tests to check that oxDNA works (not fully implemented yet) USAGE ----- oxDNA input_file OUTPUT FILE ----------- The energy.dat (default name, can be changed in the configuration file) has this layout for MD: time potential_energy kinetic_energy total_energy The energy.dat (default name, can be changed in the configuration file) has this layout for MC: time potential_energy hydrogen_bonding_energy acc_trasl acc_rot [acc_volume] Mind that potential, kinetic and total energies are divided by the number of particles. Configurations are saved in the trajectory file (appended one after the other). OBSERVABLES ----------- The observable infrastructure was devised to help building customized output from oxDNA (and DNAAnalysis) without having to dive in the simulation code itself. The relevant keys in the input file are analysis_data_output_* and data_output_* (see below). These take as an argument, between curly brackets, a series of lines that is interpreted as an input file. An example is: data_output_1 = { name = caca.dat print_every = 100 col_1 = { type = potential_energy } col_2 = { type = step units = MD } } this will print in caca.dat two colums, the first with the potential energy and the second with the steps in MD units (dt aware). The lines in between curly brackets are interpreted as input files by the single observables. See also the doxygen documentation for Observables/ObservableOutput.h oxDNA provides a plugin infrastructure to manage additional Observables. See the doxygen documentation for PluginManagement/PluginManager.h INPUT FILE OPTIONS ------------------ As always in UNIX environments, everything is case sensitive. The options are in the form key = value. Key-value pairs should be either on different lines or separated by a semicolon. There can be arbitrary spaces before and after both key and value. Line with a leading # will be treated as comments. The expected type of the value is specified between < and >. A <bool> key can be specified with values 1|0, yes|no or true|false. The | symbol (pipe) is used to indicate the different values that can be used to specify a value for the key. Keys between [ and ] are optional. Core options: T = <float> temperature of the simulation. It can be expressed in simulation units or kelvin (append a k or K after the value) or celsius (append a c or C after the value). [fix_diffusion = <bool>] if true, particles that leave the simulation box are brought back in via periodic boundary conditions. Defaults to true. [seed = <int>] seed for the random number generator. On Unix systems, defaults to either a number from /dev/urandom or to time(NULL) [confs_to_skip = <int>] how many configurations should be skipped before using the next one as the initial configuration, defaults to 0 restart_step_counter = <boolean>/<bool> false means that the step counter will start from the value read in the configuration file, true means that the step counter will start from 0/if True oxDNA will reset the step counter to 0, otherwise it will start from the step counter found in the initial configuration. Defaults to False. [external_forces = <bool>] specifies whether there are external forces acting on the nucleotides or not. If it is set to 1, then a file which specifies the external forces' configuration has to be provided (see external_forces_file) [external_forces_file = <path>] specifies the file containing all the external forces' configurations. Currently there are six supported force types: string, twist, trap, repulsion_plane, repulsion_plane_moving and mutual_trap (see EXAMPLES/TRAPS for some examples) [back_in_box = <bool>] whether particles should be brought back into the box when a configuration is printed or not, defaults to false [lastconf_file = <path>] path to the file where the last configuration will be dumped trajectory_file = <path> path to the file which will contain the output trajectory of the simulation [binary_initial_conf = <bool>] whether the initial configuration is a binary configuration or not, defaults to false [lastconf_file_bin = <path>] path to the file where the last configuration will be printed in binary format, if not specified no binary configurations will be printed [print_reduced_conf_every = <int>] every how many time steps configurations containing only the centres of mass of the strands should be printed. If 0, no reduced configurations will be printed [reduced_conf_output_dir = <path>] path to the folder where reduced configurations will be printed [no_stdout_energy = <bool>] if true oxDNA will not print the default simulation output, including the energy, to stdout. Defaults to false [print_timings = <bool>] whether oxDNA should print out to a file performance timings at the end of the simulation or not, defaults to false [timings_filename = <path>] path to the file where timings will be printed [output_prefix = <string>] the name of all output files will be preceded by this prefix, defaults to an empty string [print_input = <bool>] make oxDNA write the input key=value pairs used by the simulation in a file named input.pid, with pid being the oxDNA pid. Defaults to False. conf_file = <string> path to the starting configuration steps = <int> length of the simulation, in time steps [equilibration_steps = <int>] number of equilibration steps. During equilibration, oxDNA does not generate any output. Defaults to 0 time_scale = linear/log_lin a linear time_scale will make oxDNA print linearly-spaced configurations. a log_lin will make it print linearly-spaced cycles of logarithmically-spaced configurations. print_conf_interval = <int> if the time scale is linear, this is the number of time steps between the outputing of configurations, otherwise this is just the first point of the logarithmic part of the log_lin time scale print_conf_ppc = <int> mandatory only if time_scale == log_line. This is the number of printed configurations in a single logarithmic cycle. [print_energy_every = <int>] number of time steps between the outputing of the energy (and of the other default observables such as acceptance ratios in Monte Carlo simulations). Defaults to 0. verlet_skin = <float> width of the skin that controls the maximum displacement after which Verlet lists need to be updated. [list_type = verlet|cells|no] Type of neighbouring list to be used in CPU simulations. 'no' implies a O(N^2) computational complexity. Defaults to verlet. particle = <int> particle to apply the force to. -1 applies it to all particles. F0 = <float> Initial force. wait_time = <float> time interval over which the force is constant. Units are (MD/MC) steps. increment = <float> amount by which to increment the force every wait_time steps. stiff = <float> stiffness of the spring r0 = <float> equilibrium elongation of the spring com_list = <string> comma-separated list containing the ids of all the particles whose centre of mass is subject to the force ref_list = <string> comma-separated list containing the ids of all the particles whose centre of mass is the reference point for the force acting on the other group of particles rate = <float> growth rate of the force. It is [oxDNA energy units / (oxDNA distance units * (MD/MC) steps]. name = <string> name of the output stream. stdout or stderr are accepted values print_every = <integer> frequency of output, in number steps for oxDNA, in number of configurations for DNAnalysis [start_from = <integer>] start outputing from the given step, defaults to 0 [stop_at = <integer>] stop outputing at this step, defaults to -1 (which means never) [only_last = <bool>] if true, the output will not be appended to the stream, but it will overwrite the previous output each time, defaults to false [binary = <bool>] if true, the output will be printed in binary, defaults to false [linear = <bool>] if true the OutputObservable will save in linear scale, otherwise will use the logline scale by FS. Defaults to true [update_name_with_time = <bool>] if true the output filename will be changed by using the 'name' key as a prefix and the current step as a suffix. Defaults to false col_<n> = { type = name of the first observable [other observable options as lines of 'key = value'] } this syntax specifies the column of the output file. Note that <n> is the column index and should start from 1 ------------------------------------------------------------------------------- MD options: backend = CPU For CPU FFS backend_precision = <any> CPU FFS may use any precision allowed for a normal CPU MD simulation sim_type = FFS_MD This must be set for an FFS simulation [thermostat = no|refresh|brownian|langevin|srd] Select the simulation thermostat for MD simulations. 'no' means constant-energy simulations. 'refresh' is the Anderson thermostat. 'brownian' is an Anderson-like thermostat that refreshes momenta of randomly chosen particles. 'langevin' implements a regular Langevin thermostat. 'srd' is an (experimental) implementation of a stochastic rotational dynamics algorithm. 'no' and 'brownian' are also available on CUDA. Defaults to 'no'. gamma_trans = <float> translational damping coefficient for the Langevin thermostat. Either this or diff_coeff should be specified in the input file. newtonian_steps = <int> number of integration timesteps after which momenta are refreshed pt = <float> probability of refreshing the momenta of each particle diff_coeff = <float> base diffusion coefficient. Either pt or diff_coeff should be specified in the input file ------------------------------------------------------------------------------- MC options: ensemble = nvt|npt ensemble of the simulation [check_energy_every = <float>] oxDNA will compute the energy from scratch, compare it with the current energy and throw an error if the difference is larger then check_energy_threshold. Defaults to 10. [check_energy_threshold = <float>] threshold for the energy check. Defaults to 1e-2f for single precision and 1e-6 for double precision. delta_translation = <float> controls the trial translational displacement, which is a randomly chosen number between -0.5*delta and 0.5*delta for each direction. delta_rotation = <float> controls the angular rotational displacement, given by a randomly chosen angle between -0.5*delta and 0.5*delta radians. delta_volume = <float> controls the volume change in npt simulations. P = <float> the pressure of the simulation. Used only if ensemble == npt. [adjust_moves = <bool>] if true, oxDNA will run for equilibration_steps time steps while changing the delta of the moves in order to have an optimal acceptance ratio. It does not make sense if equilibration_steps is 0 or not given. Defaults to false [maxclust = <int>] Default: N; maximum number of particles to be moved together. Defaults to the whole system [small_system = <bool>] Default: false; whether to use an interaction computation suited for small systems. [preserve_topology = <bool>] Default: false; sets a maximum size for the move attempt to 0.5, which guarantees that the topology of the system is conserved. Also prevents very large moves and might speed up simulations of larger systems, while suppressing diffusion [umbrella_sampling = <bool>] Default: false; whether to use umbrella sampling [op_file = <string>] Mandatory if umbrella_sampling is set to true; path to file with the description of the order parameter [weights_file = <string>] Mandatory if umbrella_sampling is set to true; path to file with the weights to use in umbrella sampling [last_hist_file = <string>] Optional if umbrella_sampling is set to true, otherwise ignored; Default: last_hist.dat; path to file where the histograms associated with umbrella sampling will be stored. This is printed with the same frequency as the energy file. Should become an observable sooner or later [traj_hist_file = <string>] Optional if umbrella_sampling is set to true, otherwise ignored; Default: traj_hist.dat; path to file where the series histograms associated with umbrella sampling will be stored, allowing to monitor the time evolution of the histogram and possibly to remove parts of the simulation. This is printed with the same frequency as the energy file. Should become an observable sooner or later [init_hist_file = <string>] Optional if umbrella_sampling is set to true, otherwise ignored; Default: none; path to a file to load a previous histogram from, useful if one wants to continue a simulation to obtain more statistics. [extrapolate_hist = <float>,<float>,..., <float>] Optional if umbrella_sampling is set to true, otherwise ignored; Default: none; series of temperatures to which to extrapolate the histograms. They can be given as float in reduced units, or the units can be specified as in the T option [safe_weights = <bool>] Default: true; whether to check consistency in between order parameter file and weight file. Only used if umbrella_sampling = true [default_weight = <float>] Default: none; mandatory if safe_weights = true; default weight for states that have no specified weight assigned from the weights file [skip_hist_zeros = <bool>] Default: false; Wether to skip zero entries in the traj_hist file [equilibration_steps = <int>] Default: 0; number of steps to ignore to allow for equilibration ------------------------------------------------------------------------------- Interactions/RNAInteraction_relax.h options: relax_type = <string> Possible values: constant_force, harmonic_force; Relaxation algorithm used relax_strength = <float> Force constant for the replacement of the FENE potential Interactions/DNAInteraction_relax.h options: relax_type = <string> Possible values: constant_force, harmonic_force; Relaxation algorithm used relax_strength = <float> Force constant for the replacement of the FENE potential Interactions/DirkInteraction2.h options: length = <float> lenght of the cylinders DHS_radius = <float> radius of the diploar hard sphere on top of each cylinder DHS_rcut = <float> distance cutoff for the reaction field treatment DHS_eps = <float> background dielectric constant for the reaction field treatment Interactions/RNAInteraction2.h options: [use_average_seq = <boolean>] defaults to yes [seq_dep_file = <string>] sets the location of the files with sequence-dependent parameters [external_model = <string>] overrides default constants for the model, set in rna_model.h), by values specified by this option [salt = <float>] sets the salt concentration in M, defaults to 1 [mismatch_repulsion = <boolean>] defaults to no Interactions/DirkInteractionSin.h options: length = <float> lenght of the cylinders DHS_radius = <float> radius of the diploar hard sphere on top of each cylinder DHS_rcut = <float> distance cutoff for the reaction field treatment DHS_eps = <float> background dielectric constant for the reaction field treatment Interactions/HardCylinderInteraction.h options: height = <float> cylinder length Interactions/DNAInteraction.h options: [use_average_seq = <boolean>] defaults to yes [hb_multiplier = <float>] HB interaction multiplier applied to all the nucleotides having a custom numbered base whose magnitude is > 300, defaults to 1.0 Interactions/LJInteraction.h options: LJ_rcut = <float> interaction cutoff [LJ_kob_andersen = <bool>] Simulate a Kob-Andersen mixture. Defaults to false. [LJ_n = <int>] Generalised LJ exponent. Defaults to 6, which is the classic LJ value. Interactions/TSPInteraction.h options: TSP_rfene = <float> FENE length constant for bonded interactions TSP_sigma[type] = <float> particle diameter associated to each interaction TSP_epsilon[type] = <float> energy scale associated to each interaction TSP_attractive[type] = <float> whether the interaction contains an attractive tail or not TSP_n[type] = <int> exponent for the generalised LJ potential for each interaction [TSP_attractive_anchor = <bool>] set to true if you want the anchor monomer to be of type B instead of type A. Defaults to false [TSP_only_chains = <bool>] if true the system will be composed of chains only. The topology will be interpreted accordingly by ignoring the first value of each line (which, in the case of TSPs, is the number of arms). Defaults to false [TSP_only_intra = <bool>] if true monomers belonging to different stars will not interact. Defaults to false Interactions/TEPInteraction.h options: [use_average_seq = <boolean>] defaults to yes [hb_multiplier = <float>] HB interaction multiplier applied to all the nucleotides having a custom numbered base whose magnitude is > 300, defaults to 1.0 Interactions/DHSInteraction.h options: DHS_eps = <float> background dielectrci constant for reaction field treatment DHS_rcut = <float> cutoff for the reaction field treatment Interactions/DirkInteractionBias.h options: length = <float> lenght of the cylinders DHS_radius = <float> radius of the diploar hard sphere on top of each cylinder DHS_rcut = <float> distance cutoff for the reaction field treatment DHS_eps = <float> background dielectric constant for the reaction field treatment Interactions/RNAInteraction.h options: [use_average_seq = <boolean>] defaults to yes [seq_dep_file = <string>] sets the location of the files with sequence-dependent parameters [external_model = <string>] overrides default constants for the model, set in rna_model.h), by values specified by this option Interactions/PatchyInteraction.h options: PATCHY_N = <int> number of patches [PATCHY_N_B = <int>] number of patches on species B [PATCHY_alpha = <float>] width of patches, defaults to 0.12 Interactions/DNA2Interaction.h options: salt_concentration = <float> sets the salt concentration in M [dh_lambda = <float>] the value that lambda, which is a function of temperature (T) and salt concentration (I), should take when T=300K and I=1M, defaults to the value from Debye-Huckel theory, 0.3616455 [dh_strength = <float>] the value that scales the overall strength of the Debye-Huckel interaction, defaults to 0.0543 [dh_half_charged_ends = <bool>] set to false for 2N charges for an N-base-pair duplex, defaults to 1 Interactions/InteractionFactory.h options: [interaction_type = DNA|RNA|HS|LJ|patchy|TSP|DNA_relax|DNA_nomesh|Box|HardCylinder|HardSpheroCylinder|DHS|Dirk] Particle-particle interaction of choice. Check the documentation relative to the specific interaction for more details. Defaults to dna. Interactions/BoxInteraction.h options: box_sides = <float>, <float>, <float> sides of the box Interactions/HardSpheroCylinderInteraction.h options: length = <float> length of the spherocylinder Interactions/DirkInteraction.h options: length = <float> lenght of the cylinders DHS_radius = <float> radius of the diploar hard sphere on top of each cylinder DHS_rcut = <float> distance cutoff for the reaction field treatment DHS_eps = <float> background dielectric constant for the reaction field treatment ------------------------------------------------------------------------------- CUDA options: backend = CUDA For CUDA FFS -- NB unlike the CPU implementation, the CUDA implementation does not print extra columns with the current order parameter values whenever the energy is printed backend_precision = mixed CUDA FFS is currently only implemented for mixed precision sim_type = FFS_MD This must be set for an FFS simulation order_parameters_file = <string> path to the order parameters file ffs_file = <string> path to the file with the simulation stopping conditions. Optionally, one may use 'master conditions' (CUDA FFS only), which allow one to more easily handle very high dimensional order parameters. See the EXAMPLES/CUDA_FFS/README file for more information [ffs_generate_flux = <bool>] CUDA FFS only. Default: False; if False, the simulation will run until a stopping condition is reached; if True, a flux generation simulation will be run, in which case reaching a condition will cause a configuration to be saved but will not terminate the simulation. In the stopping condition file, the conditions must be labelled forward1, forward2, ... (for the forward conditions); and backward1, backward2, ... (for the backward conditions), ... instead of condition1, condition2, ... . To get standard flux generation, set the forward and backward conditions to correspond to crossing the same interface (and use conditions corresponding to different interfaces for Tom's flux generation). As with the single shooting run mode, the name of the condition crossed will be printed to stderr each time. [gen_flux_save_every = <integer>] CUDA FFS only. Mandatory if ffs_generate_flux is True; save a configuration for 1 in every N forward crossings [gen_flux_total_crossings = <integer>] CUDA FFS only. Mandatory if ffs_generate_flux is True; stop the simulation after N crossings achieved [gen_flux_conf_prefix = <string>] CUDA FFS only. Mandatory if ffs_generate_flux is True; the prefix used for the file names of configurations corresponding to the saved forward crossings. Counting starts at zero so the 3rd crossing configuration will be saved as MY_PREFIX_N2.dat [gen_flux_debug = <bool>] CUDA FFS only. Default: False; In a flux generation simulation, set to true to save backward-crossing configurations for debugging [check_initial_state = <bool>] CUDA FFS only. Default: False; in a flux generation simulation, set to true to turn on initial state checking. In this mode an initial configuration that crosses the forward conditions after only 1 step will cause the code to complain and exit. Useful for checking that a flux generation simulation does not start out of the A-state [die_on_unexpected_master = <bool>] CUDA FFS only. Default: False; in a flux generation simulation that uses master conditions, set to true to cause the simulation to die if any master conditions except master_forward1 or master_backward1 are reached. Useful for checking that a flux generation simulation does not enter any unwanted free energy basins (i.e. other than the initial state and the desired final state) [unexpected_master_prefix = <string>] CUDA FFS only. Mandatory if die_on_unexpected_master is True; the prefix used for the file names of configurations corresponding to reaching any unexpected master conditions (see die_on_unexpected_master). [CUDA_device = <int>] CUDA-enabled device to run the simulation on. If it is not specified or it is given a negative number, a suitable device will be automatically chosen. [CUDA_sort_every = <int>] sort particles according to a 3D Hilbert curve every CUDA_sort_every time steps. This will greatly enhnance performances for some types of interaction. Defaults to 0, which disables sorting. [threads_per_block = <int>] Number of threads per block on the CUDA grid. defaults to 2 * the size of a warp. [CUDA_list = no|verlet] Neighbour lists for CUDA simulations. Defaults to 'no'. ------------------------------------------------------------------------------- Analysis options: [analysis_confs_to_skip = <int>] number of configurations that should be excluded from the analysis. analysis_data_output_<n> = { ObservableOutput } specify an analysis output stream. <n> is an integer number and should start from 1. The setup and usage of output streams are documented in the ObservableOutput class. ------------------------------------------------------------------------------- Observables/Step.h options: [units = steps|MD] units to print the time on. time in MD units = steps * dt, defaults to step Observables/DensityProfile.h options: max_value = <float> anything with a relevant coordinate grater than this will be ignored. Mind that the observable is PBC-aware. bin_size = <float> the bin size for the profile axis = <char> Possible values: x, y, z the axis along which to compute the profile Observables/Pitch.h options: bp1a_id = <int> base pair 1 particle a id bp1b_id = <int> base pair 1 particle b id bp2a_id = <int> base pair 2 particle a id bp2b_id = <int> base pair 2 particle b id Observables/ParticlePosition.h options: particle_id = <int> particle id [orientation = <bool>] defaults to false. If 1, it also prints out the orientation [absolute = <bool>] defaults to false. If 1, does not use periodic boundaries and it prints out the absolute position of the center of mass Observables/SaltExtrapolation.h options: salts = <float>, <float>, ... list of salt concentration to extrapolate to temps = <T>, <T>, ... list of temperatures to extrapolate to, separated with commas. Temperatures can be specified in reduced units, Kelvin, Celsius as 0.10105, 30C, 30c, 30 c, 303.15 k, 303.15K, 303.15k [op_file = <string>] order parameter file. If not found, it will use the one from the input file [weights_file = <string>] weights file. If not found, the one from the input file will be used. Observables/Pressure.h options: type = pressure an observable that computes the osmotic pressure of the system [stress_tensor = <bool>] if true, the output will contain 7 fields, with the first being the total pressure and the other 6 the six independent components of the stress tensor, xx, yy, zz, xy, xz, yz Observables/HBEnergy.h options: [pairs_file = <string>] OrderParameter file containing the list of pairs whose HB energy is to be computed [base_file = <string>] file containing a list of nucleotides whose HB energy is to be computed, one nucleotide per line Observables/Distance.h options: particle_1 = <int> index of the first particle particle_2 = <int> index of the second particle. The distance is returned as r(2) - r(1) [PBC = <bool>] Whether to honour PBC. Defaults to True [dir = <float>, <float>, <float>] vector to project the distance along. Beware that it gets normalized after reading. Defaults to (1, 1, 1) / sqrt(3) Observables/ForceEnergy.h options: [print_group = <string>] limits the energy computation to the forces belonging to a specific group of forces. This can be set by adding a group_name option to each force's input. By default ForceEnergy computes the energy due to all the forces. Observables/PotentialEnergy.h options: [split = <bool>] defaults to false, it tells the observable to print all the terms contributing to the potential energy Observables/CoaxVariables.h options: particle1_id = <int> particle 1 id particle2_id = <int> particle 2 id Observables/ObservableOutput.h options: name = <string> name of the output stream. stdout or stderr are accepted values print_every = <integer> frequency of output, in number steps for oxDNA, in number of configurations for DNAnalysis [start_from = <integer>] start outputing from the given step, defaults to 0 [stop_at = <integer>] stop outputing at this step, defaults to -1 (which means never) [only_last = <bool>] if true, the output will not be appended to the stream, but it will overwrite the previous output each time, defaults to false [binary = <bool>] if true, the output will be printed in binary, defaults to false [linear = <bool>] if true the OutputObservable will save in linear scale, otherwise will use the logline scale by FS. Defaults to true [update_name_with_time = <bool>] if true the output filename will be changed by using the 'name' key as a prefix and the current step as a suffix. Defaults to false col_<n> = { type = name of the first observable [other observable options as lines of 'key = value'] } this syntax specifies the column of the output file. Note that <n> is the column index and should start from 1 Observables/Rdf.h options: max_value = <float> maximum r to consider bin_size = <float> bin size for the g(r) [axes = <string>] Possible values: x, y, z, xy, yx, zy, yz, xz, zx. Those are the axes to consider in the computation. Mind that the normalization always assumes 3D sytems for the time being. Observables/MeanVectorCosine.h options: chain_id = <int> chain id first_particle_position = <int> defaults to 0. position along the chain of the first particle on which to compute the vector's cosine with the next particle last_particle_position = <int> defaults to N-2, where N is the number of elements of the chain. Position along the chain of the last particle over which to compute the vector's cosine with the next particle vector_to_average = <int> defaults to 1. Can be 1,2, or 3 depending on the vectors we wish to consider, or 0. In that case it measures the quantity (v2*v2')(v3*v3') - |v2 ^ v2||v3 ^ v3| Observables/PairEnergy.h options: particle1_id = <int> particle 1 id particle2_id = <int> particle 2 id Observables/Configurations/PdbOutput.h options: [back_in_box = <bool>] Default: true; if true the particle positions will be brought back in the box [show = <int>,<int>,...] Default: all particles; list of comma-separated indexes of the particles that will be shown. Other particles will not appear [hide = <int>,<int>,...] Default: no particles; list of comma-separated indexes of particles that will not be shown [ref_particle = <int>] Default: -1, no action; The nucleotide with the id specified (starting from 0) is set at the centre of the box. Overriden if ref_strands is specified. Ignored if negative or too large for the system. [ref_strand = <int>] Default: -1, no action; The strand with the id specified (starts from 1) is set at the centre of the box. Ignored if negative or too large for the system. Observables/Configurations/TclOutput.h options: [back_in_box = <bool>] Default: true; if true the particle positions will be brought back in the box [show = <int>,<int>,...] Default: all particles; list of comma-separated indexes of the particles that will be shown. Other particles will not appear [hide = <int>,<int>,...] Default: no particles; list of comma-separated indexes of particles that will not be shown [print_labels = <bool>] Default: false; if true labels with the strand id are printed next to one end of the strand. [resolution = <int>] Default: 20; resolution set in the tcl file. [ref_particle = <int>] Default: -1, no action; The nucleotide with the id specified (starting from 0) is set at the centre of the box. Overriden if ref_strands is specified. Ignored if negative or too large for the system. [ref_strand = <int>] Default: -1, no action; The strand with the id specified, starting from 1, is set at the centre of the box. Ignored if negative or too large for the system. Observables/Configurations/Configuration.h options: [back_in_box = <bool>] if true the particle positions will be brought back in the box, defaults to false [show = <int>,<int>,...] list of comma-separated particle indexes whose positions will be put into the final configuration [hide = <int>,<int>,...] list of comma-separated particle indexes whose positions won't be put into the final configuration [reduced = <bool>] if true only the strand centres of mass will be printed, defaults to false Observables/Configurations/ChimeraOutput.h options: [colour_by_sequece = <bool>] Default: false; whether to coulour the bases according to the base type (A, C, G, T ------------------------------------------------------------------------------- Forward Flux Sampling (FFS) options: backend = CPU/CUDA For CPU FFS/For CUDA FFS -- NB unlike the CPU implementation, the CUDA implementation does not print extra columns with the current order parameter values whenever the energy is printed backend_precision = <any>/mixed CPU FFS may use any precision allowed for a normal CPU MD simulation/CUDA FFS is currently only implemented for mixed precision sim_type = FFS_MD This must be set for an FFS simulation order_parameters_file = <string> path to the order parameters file ffs_file = <string> path to the file with the simulation stopping conditions. Optionally, one may use 'master conditions' (CUDA FFS only), which allow one to more easily handle very high dimensional order parameters. See the EXAMPLES/CUDA_FFS/README file for more information [ffs_generate_flux = <bool>] CUDA FFS only. Default: False; if False, the simulation will run until a stopping condition is reached; if True, a flux generation simulation will be run, in which case reaching a condition will cause a configuration to be saved but will not terminate the simulation. In the stopping condition file, the conditions must be labelled forward1, forward2, ... (for the forward conditions); and backward1, backward2, ... (for the backward conditions), ... instead of condition1, condition2, ... . To get standard flux generation, set the forward and backward conditions to correspond to crossing the same interface (and use conditions corresponding to different interfaces for Tom's flux generation). As with the single shooting run mode, the name of the condition crossed will be printed to stderr each time. [gen_flux_save_every = <integer>] CUDA FFS only. Mandatory if ffs_generate_flux is True; save a configuration for 1 in every N forward crossings [gen_flux_total_crossings = <integer>] CUDA FFS only. Mandatory if ffs_generate_flux is True; stop the simulation after N crossings achieved [gen_flux_conf_prefix = <string>] CUDA FFS only. Mandatory if ffs_generate_flux is True; the prefix used for the file names of configurations corresponding to the saved forward crossings. Counting starts at zero so the 3rd crossing configuration will be saved as MY_PREFIX_N2.dat [gen_flux_debug = <bool>] CUDA FFS only. Default: False; In a flux generation simulation, set to true to save backward-crossing configurations for debugging [check_initial_state = <bool>] CUDA FFS only. Default: False; in a flux generation simulation, set to true to turn on initial state checking. In this mode an initial configuration that crosses the forward conditions after only 1 step will cause the code to complain and exit. Useful for checking that a flux generation simulation does not start out of the A-state [die_on_unexpected_master = <bool>] CUDA FFS only. Default: False; in a flux generation simulation that uses master conditions, set to true to cause the simulation to die if any master conditions except master_forward1 or master_backward1 are reached. Useful for checking that a flux generation simulation does not enter any unwanted free energy basins (i.e. other than the initial state and the desired final state) [unexpected_master_prefix = <string>] CUDA FFS only. Mandatory if die_on_unexpected_master is True; the prefix used for the file names of configurations corresponding to reaching any unexpected master conditions (see die_on_unexpected_master). -------------------------------------------------------------------------------