Understanding Nonlinear Mixed Effects Modeling for Population Pharmacokinetics

Description

This shows how NONMEM(R) <http://www.iconplc.com/innovation/nonmem/> software works. Additionally, provides functions for post-run processing of NONMEM output files, generating comprehensive PDF diagnostic reports.

Details

This package explains 'First Order(FO) approximation' method, 'First Order Conditional Estimation(FOCE)' method, and 'Laplacian(LAPL)' method of NONMEM software. It also provides post-run processing functions including PDF diagnostic reports for OFV analysis, parameter estimates, prediction diagnostics, residual diagnostics, EBE analysis, input data summary, and individual PK parameter distributions.

Author(s)

Kyun-Seop Bae <k@acr.kr>

References

1. NONMEM Users guide

2. Wang Y. Derivation of various NONMEM estimation methods. J Pharmacokinet Pharmacodyn. 2007.

3. Kang D, Bae K, Houk BE, Savic RM, Karlsson MO. Standard Error of Empirical Bayes Estimate in NONMEM(R) VI. K J Physiol Pharmacol. 2012.

4. Kim M, Yim D, Bae K. R-based reproduction of the estimation process hidden behind NONMEM Part 1: First order approximation method. 2015.

5. Bae K, Yim D. R-based reproduction of the estimation process hidden behind NONMEM Part 2: First order conditional estimation. 2016.

Examples

DataAll = Theoph
colnames(DataAll) = c("ID", "BWT", "DOSE", "TIME", "DV")
DataAll[,"ID"] = as.numeric(as.character(DataAll[,"ID"]))

nTheta = 3
nEta = 3
nEps = 2

THETAinit = c(2, 50, 0.1)
OMinit = matrix(c(0.2, 0.1, 0.1, 0.1, 0.2, 0.1, 0.1, 0.1, 0.2), nrow=nEta, ncol=nEta)
SGinit = diag(c(0.1, 0.1))

LB = rep(0, nTheta) # Lower bound
UB = rep(1000000, nTheta) # Upper bound

FGD = deriv(~DOSE/(TH2*exp(ETA2))*TH1*exp(ETA1)/(TH1*exp(ETA1) - TH3*exp(ETA3))*
             (exp(-TH3*exp(ETA3)*TIME)-exp(-TH1*exp(ETA1)*TIME)),
            c("ETA1","ETA2","ETA3"),
            function.arg=c("TH1", "TH2", "TH3", "ETA1", "ETA2", "ETA3", "DOSE", "TIME"),
            func=TRUE, hessian=TRUE)
H = deriv(~F + F*EPS1 + EPS2, c("EPS1", "EPS2"), function.arg=c("F", "EPS1", "EPS2"), func=TRUE)

PRED = function(THETA, ETA, DATAi)
{
  FGDres = FGD(THETA[1], THETA[2], THETA[3], ETA[1], ETA[2], ETA[3], DOSE=320, DATAi[,"TIME"]) 
  Gres = attr(FGDres, "gradient")
  Hres = attr(H(FGDres, 0, 0), "gradient")
  
  if (e$METHOD == "LAPL") {
    Dres = attr(FGDres, "hessian")
    Res = cbind(FGDres, Gres, Hres, Dres[,1,1], Dres[,2,1], Dres[,2,2], Dres[,3,])
    colnames(Res) = c("F", "G1", "G2", "G3", "H1", "H2", "D11", "D21", "D22", "D31", "D32", "D33") 
  } else {
    Res = cbind(FGDres, Gres, Hres)
    colnames(Res) = c("F", "G1", "G2", "G3", "H1", "H2") 
  }
  return(Res)
}

####### First Order Approximation Method # Commented out for the CRAN CPU time
#InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB, 
#         Pred=PRED, METHOD="ZERO")
#(EstRes = EstStep())           # 4 sec
#(CovRes = CovStep())           # 2 sec
#PostHocEta() # Using e$FinalPara from EstStep()
#TabStep()    

######## First Order Conditional Estimation with Interaction Method
#InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB, 
#         Pred=PRED, METHOD="COND")
#(EstRes = EstStep())           # 2 min
#(CovRes = CovStep())           # 1 min
#get("EBE", envir=e)
#TabStep()

######## Laplacian Approximation with Interaction Method
#InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB, 
#         Pred=PRED, METHOD="LAPL")
#(EstRes = EstStep())           # 4 min
#(CovRes = CovStep())           # 1 min
#get("EBE", envir=e)
#TabStep()

EBE Correlation and Covariance Tables

Description

Prints correlation and covariance tables comparing empirical Bayes estimates (EBE) with the estimated omega matrix, including:

• Correlation of covariates and EBE

• Covariance of EBE vs omega matrix

• Ratios of Cov(EBE)/OM and Cor(EBE)/Cor(OM)

Helper function for nmw_report_ebe.

Usage

.ebe_correlation_tables(tabEta, nEta, EtaNames2, OM, VarStat, FDATA)

Arguments

EBE Covariate Analysis

Description

Performs ETA vs covariate analysis including:

• Boxplots of ETAs vs categorical covariates with ANOVA p-values

• Pair plots via EBEpair

• Grid plots of ETAs vs continuous covariates with outlier labeling and lowess smoothing

Helper function for nmw_report_ebe.

Usage

.ebe_covariate_analysis(
  tabEta,
  FDATA,
  VarStat,
  nEta,
  nID,
  IDs,
  EtaNames2,
  RunNumber,
  defpar
)

Arguments

Value

data.frame, tabEta with covariate columns appended

EBE Individual Confidence Intervals and Shrinkage

Description

Computes and exports individual ETA confidence intervals, significance flags, relative standard errors, and individual shrinkage values. Writes results to EBEALL.CSV and prints per-ETA summary tables to the PDF.

For each ETA:

• LL/UL: Lower/Upper 95% confidence limit (ETA +/- 2*SE)

• ZERO: Whether CI includes zero

• RSE: Relative standard error (SE/|ETA|)

• SHR: Individual shrinkage (SE/sqrt(omega))

Helper function for nmw_report_ebe.

Usage

.ebe_individual_ci(EtaRep, nEta, nID, OM)

Arguments

EBE Distribution Plots

Description

Generates distribution diagnostic plots for each ETA including summary statistics, histograms with density overlay, Shapiro-Wilk normality test, population shrinkage, and Q-Q plots. Helper function for nmw_report_ebe.

Usage

.ebe_plot_distributions(EtaRep, nEta, OM, defpar)

Arguments

EBE Multiple Linear Regression and Influence Diagnostics

Description

Performs multiple linear regression of each ETA on covariates and generates influence diagnostic plots including:

• Cook's Distance

• Studentized deleted residuals vs predicted values

• e^2/SSE vs hat values

• COVRATIO vs DFFITS

Helper function for nmw_report_ebe.

Usage

.ebe_regression_diagnostics(
  tabEta,
  nEta,
  EtaNames,
  EtaRep,
  VarStat,
  FDATA,
  defpar
)

Arguments

Add a Covariate Column to an Existing NONMEM dataset

Description

A new covariate column can be added to an existing NONMEM dataset.

Usage

  AddCox(nmData, coxData, coxCol, dateCol = "DATE", idCol = "ID")

Arguments

Details

It first carry forward for the missing data. If NA is remained, it carry backward.

Value

A new NONMEM dataset containing the covariate column

Author(s)

Kyun-Seop Bae <k@acr.kr>

Add Dose Number, Time of Latest Dose, and Time after Latest Dose

Description

Adds DoNo (Dosing Occasion Number), ToLD (Time of Latest Dose), and TaLD (Time after Latest Dose) columns to a NONMEM dataset.

Usage

AddDoNoTaLD(
  NMData,
  ID = "ID",
  TIME = "TIME",
  AMT = "AMT",
  II = "II",
  ADDL = "ADDL",
  MDV = "MDV"
)

Arguments

Value

data.frame with DoNo, ToLD, TaLD columns added

Add a New Page to PDF

Description

Creates a new page in the PDF output with optional headers and footers.

Usage

AddPage(
  Cex = 0.8,
  Header1 = "",
  Header2 = "",
  Header3 = "",
  Footer1 = "",
  Footer2 = "",
  Footer3 = "",
  PrintRowNum = FALSE,
  StartRowNum = 1
)

Arguments

Check if All Values are NA

Description

Check if All Values are NA

Usage

AllNA(Column)

Arguments

Value

logical

Check if All Values are the Same

Description

Check if All Values are the Same

Usage

AllSame(Column)

Arguments

Value

logical

Extract Lines Between XML Tags

Description

Extract Lines Between XML Tags

Usage

BtwTagLines(Tag1, Tag2, RawRead)

Arguments

Value

character vector of lines between tags

Extract Matrix from NONMEM XML Output

Description

Parses NONMEM XML output to extract omega or sigma matrices.

Usage

BtwTagMat(Tag, RawRead, nRow)

Arguments

Value

numeric matrix

Extract Value Between XML Tags

Description

Extract Value Between XML Tags

Usage

BtwTagVal(Tag1, Tag2, RawRead)

Arguments

Value

character, the value between tags

Extract Vector of Values Between XML Tags

Description

Parses NONMEM XML output to extract theta, omega, or sigma vectors.

Usage

BtwTagVals(Tag, RawRead)

Arguments

Value

numeric vector of values

Calculate TaLD (Time after Latest Dose) for Report Generation

Description

Calculates dosing occasion information for each record in NONMEM data. Handles ADDL/II dose expansion when those columns are present. This is a shared helper used by report_pred.R and report_resid.R to avoid code duplication.

Usage

CalcTaLDForReport(FDATA, nRec)

Arguments

Value

list with components:

ToLD

numeric vector, time of latest dose for each record

TaLD

numeric vector, time after latest dose for each record

DOCC

numeric vector, dosing occasion count for each record

NewID

numeric vector, ID with fractional dosing occasion (ID + DOCC/100)

Close PDF Output

Description

Closes the current PDF device.

Usage

ClosePDF()

Combine the demographics(DM), dosing(EX), and DV(PC) tables into a new NONMEM dataset

Description

A new NONMEM dataset can be created from the demographics, dosing, and DV tables.

Usage

  CombDmExPc(dm, ex, pc)

Arguments

Details

Combining a demographics, a dosing, and a concentration table can produce a new NONMEM dataset.

Value

A new NONMEM dataset

Author(s)

Kyun-Seop Bae <k@acr.kr>

Connect Two Points with Lines

Description

Draws connecting lines between parent and child nodes in flow diagram.

Usage

ConnPoint(pt1, pt2)

Arguments

Count THETA, ETA, EPS Parameters from EXT Data Frame

Description

Counts the number of THETA, OMEGA, and SIGMA parameters from column names of a NONMEM .ext file data frame. Derives the number of ETAs and EPSILONs from the lower-triangular element count using the formula n = (sqrt(1 + 8k) - 1) / 2.

Usage

CountEXTParams(EXT)

Arguments

Details

This function replaces duplicated parameter counting loops that were previously repeated in report_ebe.R, report_ofv.R, and report_param.R.

Value

list with components:

nTheta

integer, number of THETA columns

nEta

integer, number of ETAs (derived from OMEGA elements)

nEps

integer, number of EPSILONs (derived from SIGMA elements)

Covariance Step

Description

It calculates standard errors and various variance matrices with the e$FinalPara after estimation step.

Usage

CovStep()

Details

Because EstStep uses nonlinear optimization, covariance step is separated from estimation step. It calculates variance-covariance matrix of estimates in the original scale.

Value

Author(s)

Kyun-Seop Bae <k@acr.kr>

References

NONMEM Users Guide

See Also

EstStep, InitStep

Examples

# Only after InitStep and EstStep
#CovStep()

Diagnostic Plot for Post-Processing

Description

Creates spaghetti-style diagnostic plots with individual ID labels.

Usage

DxPlotPost(x, y, mat, xlbl, ylbl, smooth, xlm = "", ylm = "", Log = "")

Arguments

EBE Pair Plot

Description

Creates pairs plot of ETAs vs covariates with histograms and correlations.

Usage

EBEpair(tabEta, RunNumber, each = FALSE)

Arguments

Estimation Step

Description

This estimates upon the conditions with InitStep.

Usage

EstStep()

Details

It does not have arguments. All necessary arguments are stored in the e environment. It assumes "INTERACTION" between eta and epsilon for "COND" and "LAPL" options. The output is basically same to NONMEM output.

Value

Author(s)

Kyun-Seop Bae <k@acr.kr>

References

NONMEM Users Guide

See Also

InitStep

Examples

# Only After InitStep
#EstStep()

Expand Dose History with ADDL/II Records

Description

Expands compressed dosing records (ADDL/II) into individual dose records.

Usage

ExpandDoseHist(DoseHistTab)

Arguments

Value

data.frame with expanded dose records

Check for Tag in File

Description

Check for Tag in File

Usage

FileTag(FileName, Tag)

Arguments

Value

logical or NA

Test Functional Dependency

Description

Tests whether DepNameList columns are functionally dependent on DetNameList columns in a NONMEM table.

Usage

FuncDep(NMTable, DetNameList, DepNameList)

Arguments

Value

logical, TRUE if functionally dependent

Get Corrected AIC (AICc)

Description

Get Corrected AIC (AICc)

Usage

GetAICc()

Value

numeric, the corrected AIC value

Get Count of All Thetas

Description

Get Count of All Thetas

Usage

GetCountAllTheta()

Value

integer, total number of thetas

Get Count of Epsilons

Description

Get Count of Epsilons

Usage

GetCountEps()

Value

integer, number of epsilon parameters

Get Count of Etas

Description

Get Count of Etas

Usage

GetCountEta()

Value

integer, number of eta parameters

Get Count of Fixed Thetas

Description

Get Count of Fixed Thetas

Usage

GetCountFixedTheta()

Value

integer

Get Count of Observations

Description

Get Count of Observations

Usage

GetCountObs()

Value

integer, total number of observation records

Get Count of Unfixed Omega Elements

Description

Get Count of Unfixed Omega Elements

Usage

GetCountOmega()

Value

integer

Get Count of Parameters

Description

Get Count of Parameters

Usage

GetCountPara()

Value

integer, total number of estimated parameters

Get Count of Unfixed Epsilons

Description

Get Count of Unfixed Epsilons

Usage

GetCountUnfixedEps()

Value

integer

Get Count of Unfixed Thetas

Description

Get Count of Unfixed Thetas

Usage

GetCountUnfixedTheta()

Value

integer

Get Current Model Name from Working Directory

Description

Detects model name by finding the .xml file in the current directory. Falls back to extracting from folder name if no .xml file is found.

Usage

GetCurModelName()

Value

character, the model name

Get Estimation Method

Description

Identifies the estimation method from the .ext file header.

Usage

GetEstMethod()

Value

character, abbreviation of the estimation method

Get Model Names from Folder Names

Description

Extracts model names by splitting folder names on period.

Usage

GetModelNames(vName)

Arguments

Value

character vector of model names

Get Objective Function Value

Description

Retrieves the final OFV from the .ext file in the current directory.

Usage

GetOFV()

Value

numeric, the objective function value

Get Off-Diagonal Omega Count

Description

Get Off-Diagonal Omega Count

Usage

GetOffDiagOmega()

Value

integer

Get PROBLEM Statement Value

Description

Get PROBLEM Statement Value

Usage

GetProbVal(Tag)

Arguments

Value

character, the extracted value

Get Termination Reason from PRINT.OUT

Description

Get Termination Reason from PRINT.OUT

Usage

GetReason(folderName)

Arguments

Value

character, termination reason code

Initialization Step

Description

It receives parameters for the estimation and stores them into e environment.

Usage

InitStep(DataAll, THETAinit, OMinit, SGinit, LB, UB, Pred, METHOD)

Arguments

Details

Prediction function should return not only prediction values(F or IPRED) but also G (first derivative with respect to etas) and H (first derivative of Y with respect to epsilon). For the "LAPL", prediction function should return second derivative with respect to eta also. "INTERACTION" is TRUE for "COND" and "LAPL" option, and FALSE for "ZERO". Omega matrix should be full block one. Sigma matrix should be diagonal one.

Value

This does not return values, but stores necessary values into the environment e.

Author(s)

Kyun-Seop Bae <k@acr.kr>

References

NONMEM Users Guide

Examples

DataAll = Theoph
colnames(DataAll) = c("ID", "BWT", "DOSE", "TIME", "DV")
DataAll[,"ID"] = as.numeric(as.character(DataAll[,"ID"]))

nTheta = 3
nEta = 3
nEps = 2

THETAinit = c(2, 50, 0.1) # Initial estimate
OMinit = matrix(c(0.2, 0.1, 0.1, 0.1, 0.2, 0.1, 0.1, 0.1, 0.2), nrow=nEta, ncol=nEta)
OMinit
SGinit = diag(c(0.1, 0.1))
SGinit

LB = rep(0, nTheta) # Lower bound
UB = rep(1000000, nTheta) # Upper bound

FGD = deriv(~DOSE/(TH2*exp(ETA2))*TH1*exp(ETA1)/(TH1*exp(ETA1) - TH3*exp(ETA3))*
             (exp(-TH3*exp(ETA3)*TIME)-exp(-TH1*exp(ETA1)*TIME)),
            c("ETA1","ETA2","ETA3"),
            function.arg=c("TH1", "TH2", "TH3", "ETA1", "ETA2", "ETA3", "DOSE", "TIME"),
            func=TRUE, hessian=TRUE)
H = deriv(~F + F*EPS1 + EPS2, c("EPS1", "EPS2"), function.arg=c("F", "EPS1", "EPS2"), func=TRUE)

PRED = function(THETA, ETA, DATAi)
{
  FGDres = FGD(THETA[1], THETA[2], THETA[3], ETA[1], ETA[2], ETA[3], DOSE=320, DATAi[,"TIME"])
  Gres = attr(FGDres, "gradient")
  Hres = attr(H(FGDres, 0, 0), "gradient")

  if (e$METHOD == "LAPL") {
    Dres = attr(FGDres, "hessian")
    Res = cbind(FGDres, Gres, Hres, Dres[,1,1], Dres[,2,1], Dres[,2,2], Dres[,3,])
    colnames(Res) = c("F", "G1", "G2", "G3", "H1", "H2", "D11", "D21", "D22", "D31", "D32", "D33")
  } else {
    Res = cbind(FGDres, Gres, Hres)
    colnames(Res) = c("F", "G1", "G2", "G3", "H1", "H2")
  }
  return(Res)
}

######### First Order Approximation Method
InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB,
         Pred=PRED, METHOD="ZERO")

######### First Order Conditional Estimation with Interaction Method
InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB,
         Pred=PRED, METHOD="COND")

######### Laplacian Approximation with Interaction Method
InitStep(DataAll, THETAinit=THETAinit, OMinit=OMinit, SGinit=SGinit, LB=LB, UB=UB,
         Pred=PRED, METHOD="LAPL")

Integer Variable Statistics

Description

Prints unique values for all integer variables.

Usage

IntStat(NMTable, VarStat)

Arguments

Check if Model Uses Log-Transformed DV (from Control File)

Description

Check if Model Uses Log-Transformed DV (from Control File)

Usage

LogDV(CtlFileName)

Arguments

Value

logical, TRUE if LOG(F) is found after $PRED or $ERROR

Check if Model Uses Log-Transformed DV (from FSUBS)

Description

Check if Model Uses Log-Transformed DV (from FSUBS)

Usage

LogDV2()

Value

logical, TRUE if LOG(F) is found after SUBROUTINE PRED or ERROR in FSUBS

Match End of String

Description

Match End of String

Usage

MatchEnd(vName0, End0, IgnoreCase = TRUE)

Arguments

Value

logical vector

Merge IDStat with Individual OFV from PHI

Description

Computes variable and individual statistics from NONMEM data, then merges individual OFV values from the PHI file. This is a common pattern used in report_ofv.R and report_resid.R.

Usage

MergeIDStatOFV(FDATA, PHI)

Arguments

Value

list with components:

IDStat3

data.frame, merged individual statistics with OFV

VarStat

matrix, variable statistics from NMVarStat()

IDStat

list, individual statistics from NMIDStat()

Check Minimization Success

Description

Check Minimization Success

Usage

MinSuccess()

Value

logical, TRUE if minimization was successful

Internal Min Util Functions

Description

Internal Min Util functions

Details

These are not to be called by the user.

NONMEM Individual (ID) Statistics

Description

Computes per-subject statistics including record counts, DV counts, dosing information, and sorting checks.

Usage

NMIDStat(NMTable)

Arguments

Value

list with IDStat data.frame, sortedID logical, sortedDT logical

NONMEM Variable Statistics

Description

Computes statistics for each variable in a NONMEM dataset including counts of NA, zero, positive values, unique values, and functional dependency checks.

Usage

NMVarStat(NMTable)

Arguments

Value

matrix with variable statistics

Post-Hoc OFV and CWRES Calculation

Description

Calculates individual OFV and CWRES using METHOD=COND or METHOD=ZERO.

Usage

OBJfpost(ctlname, sdtab)

Arguments

Value

list with OFV, OFVi, CWRES

OFV Report Split Screen Layout

Description

Layout matrix for the S1-OFV report's split screen display. Used with split.screen() to create overlapping panels on the first page of the OFV diagnostic report.

Format

A 4x4 numeric matrix. Each row defines a screen region as c(left, right, bottom, top) in normalized device coordinates:

Row 1

Full page (0, 1, 0, 1) for text content

Row 2

Top-right inset (0.55, 0.97, 0.6, 0.95) for scatter plot

Row 3

Middle-right inset (0.55, 0.97, 0.45, 0.8) for histogram

Row 4

Bottom-right inset (0.55, 0.97, 0.1, 0.45) for QQ plot

Internal Obj Functions

Description

Internal Obj functions

Details

These are not to be called by the user.

Draw Model Development Flow Diagram

Description

Creates a graphical model development flow diagram from SumOut results.

Usage

Outline(MDL, MDLName, out.start = "0", out.end = "zzzzzzzz", Target = "PDF")

Arguments

Character Panel for Pairs Plot

Description

Character Panel for Pairs Plot

Usage

PanelChar(x, y)

Arguments

Correlation Panel for Pairs Plot

Description

Correlation Panel for Pairs Plot

Usage

PanelCor(x, y, digits = 2, prefix = "", cex.cor)

Arguments

Histogram Panel for Pairs Plot

Description

Histogram Panel for Pairs Plot

Usage

PanelHist(x, ...)

Arguments

Parse Data Item Names from NONMEM Output

Description

Parse Data Item Names from NONMEM Output

Usage

ParseItemName(RawRead)

Arguments

Value

character vector of item names

Parse Fixed-Width Format Item Widths from NONMEM Output

Description

Parse Fixed-Width Format Item Widths from NONMEM Output

Usage

ParseItemWidth(RawRead)

Arguments

Value

list with LineCount and Widths

Parse Value After Tag from NONMEM Output

Description

Extracts the value following a tag string (after the colon) in NONMEM output lines.

Usage

ParseOut(TagString, RawRead)

Arguments

Value

character, the value after the tag

Plot Distribution Panel (Summary + Histogram + QQ)

Description

Generates a 3-panel distribution diagnostic plot consisting of:

1. Summary statistics text (min, Q1, median, mean, Q3, max, SD)

2. Histogram with kernel density overlay and Shapiro-Wilk test

3. Normal Q-Q plot

This is a common pattern shared across report_ebe.R, report_indipk.R, and report_resid.R.

Usage

PlotDistribution(
  var.data,
  var.name,
  show_shrinkage = FALSE,
  omega_var = NULL,
  show_ttest = FALSE,
  show_cv = FALSE
)

Arguments

Details

The function expects that the graphics device is already open and par() has been set for a multi-panel layout (e.g., mfrow). Each call to this function fills 3 panels (one row of a typical layout).

The Shapiro-Wilk test is only performed when the data has positive standard deviation and fewer than 5000 observations.

Prepare PDF Output

Description

Opens a PDF device for NONMEM report generation.

Usage

PrepPDF(FileName, Paper = "letter", FontFamily = "Courier")

Arguments

Print Multiple Lines of Text

Description

Prints a character vector of text lines across multiple pages as needed.

Usage

PrinMTxt(
  MTxt,
  Cex = 0.8,
  Header1 = "",
  Header2 = "",
  Header3 = "",
  Footer1 = "",
  Footer2 = "",
  Footer3 = "",
  PrintRowNum = FALSE
)

Arguments

Print Text at Position

Description

Prints text at a specific row and column position on the PDF page.

Usage

PrinTxt(Row, Col, Text, Cex = 0.8)

Arguments

Read EXT File from Current Directory

Description

Reads the NONMEM .ext file corresponding to the current model (detected by GetCurModelName()). This is a shared helper used by GetCountAllTheta(), GetCountEps(), GetCountOmega(), and GetCountUnfixedEps() to avoid repeated file-reading logic.

Usage

ReadEXTFile()

Value

data.frame of the .ext file contents, or NULL if the file is not found in the current directory.

Read FDATA.CSV with Auto-Detection of Header Format

Description

Reads FDATA.CSV, auto-detecting whether it has a NONMEM-style 2-row header (skip=2) or a standard 1-row header (skip=0).

Usage

ReadFDATA(file = "FDATA.CSV")

Arguments

Value

data.frame

Remove Rows with NA Values

Description

Remove Rows with NA Values

Usage

RemoveNA(RawData)

Arguments

Value

data.frame with rows containing NA removed

Rename Columns

Description

Rename Columns

Usage

RenCol(ColList, OldCol, NewCol)

Arguments

Value

character vector with renamed columns

Residual Count Test Using Binomial Distribution

Description

Tests whether residual counts at various Z-value thresholds follow expected binomial distributions.

Usage

ResTest(
  ResTab,
  TestCols = c("WRES", "CWRE", "IWRE"),
  ZVals = c(0, 1, 2, 3),
  PctVals = c(0.005, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5)
)

Arguments

Value

matrix of test results

Remove Columns from Table

Description

Remove Columns from Table

Usage

RmvCol(Tab, OldCols)

Arguments

Value

data.frame with specified columns removed

Remove Fixed (Non-Varying) Columns

Description

Removes columns where all values are the same across rows.

Usage

RmvFixed(Table)

Arguments

Value

data.frame with fixed columns removed

Remove Zero Rows/Cols from Symmetric Matrix

Description

Remove Zero Rows/Cols from Symmetric Matrix

Usage

RmvZero(SymmMat)

Arguments

Value

matrix with zero rows/columns removed

Check Standard Error Success

Description

Check Standard Error Success

Usage

SESuccess()

Value

logical, TRUE if standard errors were computed successfully

Summarize Multiple NONMEM Runs

Description

Scans subdirectories matching RunExt and collects summary metrics for model development flow analysis.

Usage

SumOut(FileExt = ".CTL", RunExt = ".R75", OutExt = ".OUT")

Arguments

Value

data.frame with summary of all runs

Summarize Single NONMEM Run

Description

Collects key metrics from a single NONMEM run folder.

Usage

SumOut1(ModelName)

Arguments

Value

data.frame with one row of summary metrics

Table Step

Description

This produces standard table.

Usage

TabStep()

Details

It does not have arguments. All necessary arguments are stored in the e environment. This is similar to other standard results table.

Value

A table with ID, TIME, DV, PRED, RES, WRES, derivatives of G and H. If the estimation method is other than 'ZERO'(First-order approximation), it includes CWRES, CIPREDI(formerly IPRED), CIRESI(formerly IRES).

Author(s)

Kyun-Seop Bae <k@acr.kr>

References

NONMEM Users Guide

See Also

EstStep

Examples

# Only After EstStep
#TabStep()

Trimming and beautifying NONMEM original OUTPUT file

Description

TrimOut removes unnecessary parts from NONMEM original OUTPUT file.

Usage

  TrimOut(inFile, outFile="PRINT.OUT")

Arguments

Details

NONMEM original OUTPUT file contains unnecessary parts such as CONTROL file content, Start/End Time, License Info, Print control characters such as "+", "0", "1". This function trims those.

Value

outFile will be written in the current working folder or designated folder. This returns TRUE if the process was smooth.

Author(s)

Kyun-Seop Bae <k@acr.kr>

environment for internal data

Description

This is for the storage of intermediate data.

Multiple Linear Regression with Influence Diagnostics

Description

Performs multiple linear regression and computes influence diagnostics including DFFITS, DFBETAS, Cook's Distance, COVRATIO, and R-Student.

Usage

mlr2(y, x.raw, standardize = 0)

Arguments

Value

list with model estimates and influence diagnostics

Generate EBE Diagnostics Report (S5-EBE.PDF)

Description

Generates a PDF report with empirical Bayes estimate (EBE) analysis including ETA distributions, shrinkage, covariate relationships, correlations, and multiple linear regression with influence diagnostics.

Usage

nmw_report_ebe(run_dir = getwd())

Arguments

Generate Individual PK Parameter Report (SC-IndiPK.PDF)

Description

Generates a PDF report with individual PK parameter distributions including summary statistics, histograms, and QQ plots.

Usage

nmw_report_indipk(run_dir = getwd())

Arguments

Generate Input Data Summary Report (SA-Input.PDF)

Description

Generates a PDF report with input data validation including individuals with no dosing/DV, duplicate records, and summary statistics.

Usage

nmw_report_input(run_dir = getwd())

Arguments

Generate OFV Diagnostic Report (S1-OFV.PDF)

Description

Generates a PDF report with objective function value diagnostics including total OFV, AICc, BIC, individual OFV per DV summary, and gradient analysis.

Usage

nmw_report_ofv(run_dir = getwd())

Arguments

Generate NONMEM Output Report (SB-Output.PDF)

Description

Includes the NONMEM PRINT.OUT file content in a PDF report.

Usage

nmw_report_output(run_dir = getwd())

Arguments

Generate Parameter Estimates Report (S2-Parameters.PDF)

Description

Generates a PDF report with theta, omega, and sigma parameter estimates, standard errors, confidence intervals, and significance flags.

Usage

nmw_report_param(run_dir = getwd())

Arguments

Generate Prediction Diagnostics Report (S3-Predictions.PDF)

Description

Generates a PDF report with spaghetti plots, DV vs PRED/IPRE, and individual fitting curves.

Usage

nmw_report_pred(run_dir = getwd())

Arguments

Generate Residual Diagnostics Report (S4-Residuals.PDF)

Description

Generates a PDF report with WRES, CWRES, IWRES diagnostics including scatter plots, histograms, normality tests, binomial count tests, and individual residual curves with run tests.

Usage

nmw_report_resid(run_dir = getwd())

Arguments

Run All NONMEM Post-Processing Reports

Description

Generates all diagnostic PDF reports for a NONMEM run in sequence: S1-OFV, S2-Parameters, S3-Predictions, S4-Residuals, S5-EBE, SA-Input, SB-Output, SC-IndiPK.

Usage

nmw_run(
  run_dir = getwd(),
  reports = c("ofv", "param", "pred", "resid", "ebe", "input", "output", "indipk")
)

Arguments

Examples

## Not run: 
# Generate all reports
nmw_run("/path/to/nonmem/run/directory")

# Generate only OFV and parameter reports
nmw_run("/path/to/run", reports = c("ofv", "param"))

## End(Not run)

Run Test Probability

Description

Calculates P(Run count <= r | m, n) for the runs test.

Usage

run.p(m, n, r)

Arguments

Value

numeric, probability

Runs Test for NONMEM Residuals

Description

Performs a runs test on residuals. Zeros are omitted.

Usage

run.test.nm(RES)

Arguments

Value

numeric, two-sided p-value