Type: Package
Title: Lasso-SAMBA Algorithm
Version: 2024.1.0
Maintainer: Auriane Gabaut <auriane.gabaut@inria.fr>
Description: Provides a lasso-based method for building mechanistic models using the SAMBA algorithm (Stochastic Approximation for Model Building Algorithm) (M Prague, M Lavielle (2022) <doi:10.1002/psp4.12742>). The package extends the 'Rsmlx' package (version 2024.1.0) to better handle high-dimensional data. It relies on the 'Monolix' software (version 2024R1; see (https://monolixsuite.slp-software.com/monolix/2024R1/), which must be installed beforehand.
SystemRequirements: 'Monolix' (<https://monolixsuite.slp-software.com/monolix/2024R1/>)
License: GPL (≥ 3)
Encoding: UTF-8
RoxygenNote: 7.3.2
Imports: doParallel, dplyr, ggplot2, grDevices, latex2exp, MASS, parallel, sharp, stringr
Depends: R (≥ 2.10), Rsmlx (≥ 2024.1.0), foreach
NeedsCompilation: no
Packaged: 2025-05-16 11:28:13 UTC; auria
Author: Auriane Gabaut [aut, cre], Mélanie Prague [aut], Marc Lavielle [aut] (for former code of Rsmlx and connectors)
Repository: CRAN
Date/Publication: 2025-05-20 08:30:06 UTC

LSAMBA: Lasso-SAMBA Algorithm

Description

Provides a lasso-based method for building mechanistic models using the SAMBA algorithm (Stochastic Approximation for Model Building Algorithm) (M Prague, M Lavielle (2022) doi:10.1002/psp4.12742). The package extends the 'Rsmlx' package (version 2024.1.0) to better handle high-dimensional data. It relies on the 'Monolix' software (version 2024R1; see (https://monolixsuite.slp-software.com/monolix/2024R1/), which must be installed beforehand.

Author(s)

Maintainer: Auriane Gabaut auriane.gabaut@inria.fr

Authors:

buildmlx: Automatic statistical model building

Description

buildmlx uses SAMBA (Stochastic Approximation for Model Building Algorithm), an iterative procedure to accelerate and optimize the process of model building by identifying at each step how best to improve some of the model components. This method allows to find the optimal statistical model which minimizes some information criterion in very few steps.

Usage

buildmlx(
  project = NULL,
  final.project = NULL,
  model = "all",
  prior = NULL,
  weight = NULL,
  coef.w1 = 0.5,
  paramToUse = "all",
  covToTest = "all",
  covToTransform = "none",
  center.covariate = FALSE,
  criterion = "BICc",
  linearization = FALSE,
  ll = TRUE,
  test = FALSE,
  direction = NULL,
  steps = 1000,
  n.full = 10,
  max.iter = 20,
  explor.iter = 2,
  fError.min = 0.001,
  seq.cov = FALSE,
  seq.cov.iter = 0,
  seq.corr = TRUE,
  p.max = if (buildMethod == "stepAIC") {
     0.1
 } else {
     1
 },
  p.min = c(0.075, 0.05, 0.1),
  print = TRUE,
  nb.model = 1,
  nfolds = 5,
  alpha = 1,
  nSS = 1000,
  buildMethod = "lasso",
  FDR_thr = 0.1
)

Arguments

project

a string: the initial Monolix project

final.project

a string: the final Monolix project (default adds "_built" to the original project)

model

components of the model to optimize c("residualError", "covariate", "correlation"), (default="all")

prior

list of prior probabilities for each component of the model (default=NULL)

weight

list of penalty weights for each component of the model (default=NULL)

coef.w1

multiplicative weight coefficient used for the first iteration only (default=0.5)

paramToUse

list of parameters possibly function of covariates (default="all")

covToTest

components of the covariate model that can be modified (default="all")

covToTransform

list of (continuous) covariates to be log-transformed (default="none")

center.covariate

TRUE/FALSE center the covariates of the final model (default=FALSE)

criterion

penalization criterion to optimize c("AIC", "BIC", "BICc", gamma) (default=BICc)

linearization

TRUE/FALSE whether the computation of the likelihood is based on a linearization of the model (default=FALSE)

ll

TRUE/FALSE compute the observe likelihood and the criterion to optimize at each iteration

test

TRUE/FALSE perform additional statistical tests for building the model (default=FALSE)

direction

for stepAIC method, method for covariate search c("full", "both", "backward", "forward"), (default="full" or "both")

steps

for stepAIC method, maximum number of iteration for stepAIC (default=1000)

n.full

for stepAIC method, maximum number of covariates for an exhaustive comparison of all possible covariate models (default=10)

max.iter

maximum number of iterations (default=20)

explor.iter

number of iterations during the exploratory phase (default=2)

fError.min

minimum fraction of residual variance for combined error model (default = 1e-3)

seq.cov

TRUE/FALSE whether the covariate model is built before the correlation model

seq.cov.iter

number of iterations before building the correlation model (only when seq.cov=F, default=0)

seq.corr

TRUE/FALSE whether the correlation model is built iteratively (default=TRUE)

p.max

maximum p-value used for removing non significant relationships between covariates and individual parameters (default=0.1 for stepAIC and 1 for lasso)

p.min

vector of 3 minimum p-values used for testing the components of a new model (default=c(0.075, 0.05, 0.1))

print

TRUE/FALSE display the results (default=TRUE)

nb.model

number of models to display at each iteration (default=1)

nfolds

for lasso method, number of folds (default=10)

alpha

for lasso method, the elasticnet mixing parameter, between 0 and 1. alpha=1 is the lasso penalty, alpha=0 is the ridge penalty.

nSS

for lasso method, number of resampling iterations for stability selection.

buildMethod

the method used to build the covariate model (default="lasso")

FDR_thr

for lasso method, upper-bounds in FDP of calibrated stability selection (default=0.1)

Details

For covariates model building, covariate selection can be achieved by stepAIC, as the original SAMBA algorithm was implemented in Rsmlx package (Prague and Lavielle, 2020 ; Mihaljevic, 2023) and by a lasso approach enhanced by stability selection (Bodinier et al., 2023).

Value

a new Monolix project with a new statistical model.

References

Prague M, Lavielle M. SAMBA: A novel method for fast automatic model building in nonlinear mixed-effects models. CPT Pharmacometrics Syst Pharmacol. 2022; 11: 161-172. doi:10.1002/psp4.12742

Bodinier B, Filippi S, Haugdahl Nøst T, Chiquet J, Chadeau-Hyam M. Automated calibration for stability selection in penalised regression and graphical models. Journal of the Royal Statistical Society Series C: Applied Statistics. 2023 ; 72: 1375–1393. doi:10.1093/jrsssc/qlad058

Mihaljevic F (2023). Rsmlx: R Speaks 'Monolix'. R package version2023.1.5, https://CRAN.R-project.org/package=Rsmlx.

Examples

## Not run: 
project <- getMLXdir()

res = buildmlx(project = project,
               buildMethod = "lasso",
               model='covariate',
               test=FALSE)

getIndividualParameterModel()

## End(Not run)

Get monolix demo project path.

Description

Get monolix demo project path.

Usage

getMLXdir()

Value

path to the monolix demo from 'Monolix' software

Examples

## Not run: 
print(getMLXdir())

## End(Not run)