+--------------------------------------------------------------------+
 |                                                                    |
 |                             ADVAN9_15                              |
 |                                                                    |
 +--------------------------------------------------------------------+

 MEANING: Choice of Pharmacokinetic Model for PREDPP
 CONTEXT: Option of NM-TRAN $SUBROUTINES record

 USAGE:
 $SUBROUTINES [ADVAN=]ADVAN9
 $SUBROUTINES [ADVAN=]ADVAN15
 $SUBROUTINES [ADVAN=]ADVAN17

 SAMPLE:
 $SUBROUTINE ADVAN9
 $SUBROUTINE ADVAN15
 $SUBROUTINE ADVAN17

 DISCUSSION:
 ADVAN9  and ADVAN15 and ADVAN17 are routines in PREDPP's library which
 implement the general non-linear model with equilibrium  compartments.
 This  general  non-linear model is used for systems in which a drug is
 distributed between compartments according to a system of  first-order
 differential-algebraic  processes.   It  may be useful when the system
 consists of only stiff differential equations, and there are no  equi-
 librium  compartments  (i.e., no algebraic equations).  It may also be
 used when the system consists of only algebraic  equations  (i.e.,  no
 differential equations).

 A  $MODEL  record  is  required to describe the compartments and their
 attributes.  The $DES record describes the differential equations,  if
 any.   The $AES and $AESINITIAL records describe the algebraic expres-
 sions, if any.

 TOL is required; (See $SUBROUTINES).

 TRANS routines that may be used: TRANS1

 Explicit Basic PK parameters:

      P(n) (nth basic PK parameter)

 Implicit basic PK parameters:

      PK-defined variables used also in $AES or $DES blocks

 Additional PK parameters:

 Suppose there are m compartments in the system, including  the  output
 compartment.  For each compartment n in the system (n=1, ..., m):

      Sn - Scale for nth compartment
      S0 - Alternate name for scale for output compartment

 For each dosable compartment n in the system:

      Fn - Bioavailability for nth compartment
      Rn - Rate for nth compartment
      Dn - Duration for nth compartment
      ALAGn - Absorption lag for nth compartment

 Other additional PK parameters:

      F0 - Output fraction (also called Fm, FO)
      XSCALE - Time scale
      MTIME(i) - Model event times

 With  NONMEM  7.4,  ATOL  may  be  specified  for  ADVAN9 and ADVAN15.
 (ADVAN17 is new with NONMEM 7.5. The following applies to  ADVAN17  as
 well.)   TOL (relative tolerance) and ATOL (Absolute tolerance) may be
 specified on the $SUBROUTINE record, the $TOL record, or  by  a  user-
 supplied  TOL subroutine.  See also the ATOL option of the $ESTIMATION
 and $COVARIANCE records.

 ATOL is not required.   The  default  is  12  (that  is,  accuracy  is
 10**(-12)).  Usually the problem runs quickly when using this setting.
 On occasion, however, you may want to  reduce  ATOL  (usually  set  it
 equal to that of TOL), and improve speeds of up to 3 to 4 fold.

 With  ADVAN9,  the  user  may  find that the TOL option should specify
 larger NRD values than for other ADVANs (e.g., ADVAN6).  Values  of  7
 or 8 may not be unreasonable.

 Reserved  variable MXSTEP (the  maximum  number  of integration steps)
 can be specified for ADVAN9 and ADVAN15 and ADVAN17, as  well  as  for
 ADVAN13 and ADVAN14 and ADVAN16.
 (See mxstep).

 With  ADVAN9,  if there is a period of time during which some compart-
 ment's amount should be zero, that compartment should be  turned  off.
 Otherwise, very small amounts can appear in the compartment, which can
 cause difficulties during subsequent time periods.

 By default, the initial conditions  (i.e.,  compartment  amounts)  are
 zero  at the start of each individual record. Different initial condi-
 tions may be specified using a compartment initialization block in PK.
 When  endogenous  drug is provided in the differential equations, non-
 zero initial conditions can be  computed  using  a  steady-state  dose
 event  record  with SS>0 and AMT=0 and RATE=0, or using the I_SS (Ini-
 tial Steady State) feature of MODEL and/or PK.  Note that with  either
 method,  steady-state  is  computed only for compartments specified by
 differential equations.  The equilibrium compartments are not computed
 (i.e.,  the algebraic equations are not evaluated) until the system is
 advanced beyond the initial value of TIME.
 The CALL data item may be used to force a call to  ADVAN,  which  will
 then evaluate the algebraic equations.
 (See Compartment_Initialization_Block, SS_dose).
 (See $model, $pk, model, pk).
 (See Initial_Steady_State:_I_SS,ISSMOD).

 ADVAN9  uses  the Lawrence Livermore National Laboratory (LLNL) solver
 for differential-algebraic equations, implicit form (LSODI), using the
 backward differentiation formulas (BDF) for stiff problems.

 ADVAN15  uses IDA from the LLNL SUNDIALS system for differential-alge-
 braic equations and is a descendent of  the  LSODI1  (ADVAN9)  system.
 ADVAN15  also  has  a  root-finding algorithm.  The user may choose to
 make modifications to parameters in ..\pr\IDAU.f90.
 See ..\guides\idas_guide.pdf and ..\guides\ida_guide.pdf.
 See INTRODUCTION TO NONMEM 7, $SUBROUTINES: Yet Another New  Differen-
 tial Equation Solving Method: IDAS (ADVAN15)

 ADVAN16 and ADVAN17 use the RADAR5 Delay Differential Equation Solver.
 ADVAN17 may be used when there are equilibrium compartments.
 See INTRODUCTION TO NONMEM 7,  Using  the  RADAR5  Delay  differential
 equation Solver with the ddexpand program.
 See also ..\guides\manrad5-v2.pdf.

 Required PREDPP Library subroutines:
 ADVAN9 - ADDA, FCN5, JAC, LEQT22, LSODI1, RES, ZSPOW2
 ADVAN15 - ADDA, FCN5, JAC, LEQT22, IDA, RES, ZSPOW2
 ADVAN17  -  JAC,  LEQT22,  RADAR5NM, RADAR5U, RES, ZSPOW2, FCN2, FCN4,
 FCN5

 1    Analytical second derivatives used with the Laplacian method  are
      not obtained with ADVAN9 and ADVAN15 and ADVAN17.  Numerical sec-
      ond derivatives must be used.

REFERENCES: Guide VI Section VI.B, VII.C.9
REFERENCES: Guide IV Section V.C.8


  
Go to main index.
  
Created by nmhelp2html v. 1.0 written by Niclas Jonsson (Modified by AJB 5/2006,11/2007,10/2012)