presim

SYNOPSIS

Assemble header file for FORTRAN-77 solver program

SYNTAX

    presim    -commands 

DESCRIPTION

This program is used to assemble two include files (simsize.h, dart.h) that are used in the assembly of different versions of the dart solver programs. The include files are used to dimension parameters for different arrays in the model, and to define the cpp flags for inclusion of different modules within the assembled code.

FLAGS

  • ipts nnn
    REQUIRED. The number of tissue nodes in the input mesh. The upper bounds on the allowed size will depend on the achine architecture and the specific implementation. For example, the DiFrancesco-Noble membrane model will require space for approximately
  • idf nnn
    REQUIRED. The connectivity dimensions. For regular/structured grids, set idf to 1 for a 1-D cable with 2 connections per internal node, 2 for a 2-D sheet with 4 connections per internal node and 3 for a 3-D parallelepiped with 6 connections per internal node. For higher order finite difference schemes, idf should be set to (Max number of connections at a single node)/2
  • idir HOME/include
    REQUIRED. The directory where the include file simsize.h will be placed.
  • icoars nnn
    Used only in mesh adaption simulations to specify the size of the initial coarse grid.
  • Machine-specific routines

  • cray
    Use Cray Research, Inc. scilib routines written into the code.
  • dxml
    Use Digital Equipment Corporation dxml routines written into the code.
  • pcg
    DEFAULT. Portable preconditioned conjugate gradient solver
  • Temporal integration methods for Vm

  • eul
    DEFAULT. Use the forward difference on the capacitive term to integrate via Euler's method.
  • dtar
    Primes dart to dynamically track the active region resulting in much faster runs, but loss of repolarization information. Only works with explicit integration
  • simp
    Use a semi-implicit scheme for the integration of Vm.
  • For the 1-D case, the semi-implicit finite difference stencil leads to a tridiagonal solve or standard Crank-Nicholson. Specify the -cray option in conjuction with -simp for a Cray-specific tridiagonal solver.
  • Additionally specify pcg in conjunction with hyb and imaxnz for the 2-D case and higher when running on a machine other than the cray.
  • Additionally specify cray in conjunction with hyb and imaxnz for the 2-D case and higher when running on the cray.
  • Additionally specify nspcg in conjunction with -nx, -ny, and -nz options for structured grids and use of the NSPCG solver package. In order to use this package, you must be able to link to the NPSCG library on the target machine.
  • maxnz
    NOTE: for either a bido or a bath simulation, the maximum number of non-zero entries must be specified at the presim step. Although code compilation will complete, the executable should warn if the bmaxnz value specified does not EXACTLY match the bmaxnz value needed.
  • mworki
    Integer work space size for simp and a preconditioned conjugate gradient solver. The executable will return the required value for this parameter in the event the specified value is too small.
  • mworkr
    Real work space size for simp and a preconditioned conjugate gradient solver. The executable will return the required value for this parameter in the event the specified value is too small.
  • Time-stepping methods

  • DEFAULT
    No time step adjustments are performed.
  • trl
    Time step is adjusted based on the maximum magnitude of dV/dt at any point in the mesh. Doubling or halving of dt only.
  • Membrane models

  • ej
    DEFAULT. Ebihara-Johnson sodium current combined with Spach-Kootsey leakage current.
  • fhn
    FitzHugh-Nagumo model.
  • br
    Beeler-Reuter model.
  • brej
    Beeler-Reuter model with Ebihara-Johnson sodium current.
  • dn
    DiFrancesco-Noble model.
  • lr
    Luo-Rudy Phase 1 model.
  • lr2
    Luo-Rudy Phase 2 Model
  • ipore
    DeBruin-Krassowska electroporation current
  • dnl2
    DiFrancesco-Noble model mixed with the Luo-Rudy Phase 2 model. Must be used in conjunction with -sp flag.
  • sp nnn
    For models that include a ventricular component (lr) and a Purkinje component (dn), sp is the highest node number for a Purkinje node. At sp+1, the muscle portion of the model starts.
  • Domains

  • DEFAULT
    Monodomain simulation which is analogous to the case of equal anisotropy and an insulated extracellular bounding surface to the interstitial space. Tissue and no bath.
  • bido
    Bidomain simulation which is analogous to the case of unequal anisotropy and an insulated extracellular bounding surface to the interstitial space. Tissue and no bath. By default, if you run a bidomain simulation dart attempts to write values for both Vm and phio to fan-processed files vm.lst and phio.lst.
  • nx nnn,-ny nnn,-nz nnn
    The number of nodes in x,y and z. These parameters are required if -bido is specified.
  • bath
    Bidomain simulation which is analogous to the case of unequal anisotropy and an extracellular bounding conductor connected to the interstitial space.
  • ibath nnn
    The number of bath nodes in the mesh. This parameter is required if -bath is specified.
  • bmaxnz
    NOTE: for either a bido or a bath simulation, the maximum number of non-zero entries must be specified at the presim step. Although code compilation will complete, the executable should warn if the bmaxnz value specified does not EXACTLY match the bmaxnz value needed.
  • bworki
    Integer work space size for a bido or bath run and a preconditioned conjugate gradient solver. The executable will return the required value for this parameter in the event the specified value is too small.
  • bworkr
    Real work space size for a bido or bath run and a preconditioned conjugate gradient solver. The executable will return the required value for this parameter in the event the specified value is too small.
  • ref
    specify the node number to be used as the unipolar reference for the PHIO data.
  • edgestim
    specify field stimulation with current sources on opposite edges of a 2-D grid.
  • Control and post-processing

  • peak
    Store the values for peak at each node in the file vmpeak.out.
  • apa
    Store the values for the action potential area at each node in the file apa.out.
  • ugle
    This option MUST be specified for ugle output. The default is no ugle.
  • nostat
    No statistics report.
  • dbg
    Yields the node report.
  • timing
    Explicitly times portions of the code.
  • nocntu
    By default, the executable will look for *.inp files to set initial values for all integrated variables. If these are no found, the executable will use internal values. At the end of the run, the executable will write a set of *.out files to describe state variables at the last time step. This flag prevents continuation.
  • Output of time-space information

  • nolst
    No fan processing => no vm.lst file.
  • ifnpts nnn
    to specify the number of nodes in a reduced key list to be used for fan output. For example, if you are only interested in the waveforms from every 10th node of a 100 node cable, you would load the file fanintra.inp with:
    10
    20
    .
    .
    100
    
    and the fan processing would only take place at nodes on the list. It's possible to set up a separate list for the interstitial/extracellular grid in the file faninter.inp. Here, nnn specifies the number of points in these files.
  • fbgate
    Write the B gate (LR2) to bgate.lst file.
  • fdgate
    Write the D gate (DN,LR2) to dgate.lst file.
  • ffgate
    Write the F gate (LR2) to fgate.lst file.
  • ff1gate
    Write the F1 gate (DN) to f1gate.lst file.
  • ff2gate
    Write the F2 gate (DN) to f2gate.lst file.
  • ffcagate
    Write the FCA gate (LR2) to fcagte.lst file.
  • fggate
    Write the G gate (LR2) to ggate.lst file.
  • fhgate
    Write the H gate (DN,LR2) to hgate.lst file.
  • fjgate
    Write the J gate (LR2) to jgate.lst file.
  • fmgate
    Write the M gate (DN,LR2) to mgate.lst file.
  • frggate
    Write the RG gate (DN) to rggate.lst file.
  • fxgate
    Write the X gate (DN) to xgate.lst file.
  • fygate
    Write the Y gate (DN) to ygate.lst file.
  • fiion
    Write the total ionic current to the file iion.lst file.
  • fim
    Write the total membrane current to the file im.lst file.
  • fcaidn,fcailr
    Write intracelullar calcium concentration (DN,LR2)
  • fcajsr
    Write JSR calcium concentration (LR2)
  • fcansr
    Write NSR calcium concentration (LR2)
  • ficatch
    Write T-channel calcium current (LR2)
  • ficadn,ficalr
    Write total calcium current (DN,LR2)
  • fif
    Write hyperpolarizing or funny current
  • fikdn,fiklr
    Write total IK (DN,LR2)
  • fik1dn,fik1lr
    Write IK1 (DN,LR2)
  • finadn,finalr
    Write fast sodium current (DN,LR2)
  • finacad,finacal
    Write sodium-calcium exchange current (DN,LR2)
  • finaklr
    Write sodium-potassium pump current (LR2)
  • fipore
    Write electroporation current (LR2)
  • fito
    Write transient outward current (DN)


  • Last Modification: $Id: presim.html,v 1.23 1998/02/22 22:57:39 pollard Exp $