Select Page

Release notes

Rsmlx 2023.1 – (on GitHub, May 30th 2023)

  • Made the package compatible with 2023R1 version of MonolixSuite
  • Refactored bootmlx code to handle replicates one by one
    • bootmlx now generates data sets and Monolix projects just before running parameter estimation, instead of generating all data sets before estimating parameters for all replicates
    • it is now possible to delete generated data sets and projects using the additional arguments
  • Fixed a bug when running bootstrap on a project that contains unmapped variables that can be mapped to observation IDs (generated projects would have those variables automatically mapped)
  • Running bootmlx on a project with data set that contains special characters returns a more meaningful error

Rsmlx 5.0.1 – (on GitHub and CRAN, June 15th, 2022).

  • It is now possible to introduce some prior information about the statistical model in buildmlxand buildAll.
  • Additional tests can be performed (or not) for building the statistical model using buildmlx or buildAll and the SAMBA algorithm.
  • Different criteria can be used for building the PK model with pkbuild,as well as the possibility to linearize (or not) model for computing the likelihood
  • Several bugs have been fixed.

Rsmlx 4.0.3 – (on GitHub, March 11th, 2022).

  • An error occurred with bootmlx using R Studio 2022
  • Using buildAll, package lixoftConnectors is now automatically loaded.

Rsmlx 4.0.2 – (on GitHub and CRAN, March 4th, 2022).

Several bugs have been fixed, including:

  • with bootmlx, datasets are generated but not run
  • with buildmlx, ll=F produces an error
  • with buildmlx, covariates used for stratification are not correctly managed

Rsmlx 4.0.1 – (on GitHub and CRAN, February 23rd, 2022).

  • Rsmlx 4.0.1 is compatible with MonolixSuite 2021R1. It is not compatible with previous versions of Monolix.
  • Rsmlx 4.0.1 includes two new functions for model building:
    • buildVar is designed to build the best variance model for the random effects by selecting which individual parameters vary and which ones are fixed,
    • buildAll builds the complete statistical model by iteratively calling functions buildmlx and buildVar.
  • buildmlx has been improved. It now includes the use of statistical tests to check:
    • which covariate/parameter relationship or which correlation between random effects could have been missed,
    • which covariate/parameter relationship or which correlation could be removed (using the Wald test).

Rsmlx 3.0.3 – (on GitHub, May 18th, 2021).

Rsmlx 3.0.0 – (on GitHub and CRAN, January 20th, 2021).

  • Rsmlx 3.0.0 is compatible both with Monolix 2019 and Monolix 2020

Rsmlx 2.0.5 – (on GitHub, October 28th, 2020).

  • bootmlx: parametric bootstap works properly

Rsmlx 2.0.4 – (on GitHub, June 12th, 2019).

  • buildmlx: a bug introduced inadvertently in version 2.0.3 has been fixed

Rsmlx 2.0.3 – (on GitHub, June 7th, 2019).

  • bug fixes:
    • pkpopini
    • buildmlx (bug when different names of observations in the model and in the data)

Rsmlx 2.0.2 – (on CRAN and GitHub, April 30th, 2019).

  • Rsmlx 2.0 is on CRAN!
  • This new version of Rsmlx runs with the MonolixSuite 2019R1
  • No need to load the lixoftConnectors library
  • See the install procedure.

Rsmlx 2.0.1 – (on GitHub, April 26th, 2019).

minor bug fixes

Rsmlx 2.0.0 – (on GitHub, April 19th, 2019).

Rsmlx 1.1.2 – (on GitHub, October 25th, 2018)

Rsmlx is now fully compatible with MlxConnectors of MonolixSuiteR2

Rsmlx 1.1.1 – (on GitHub, October 22nd, 2018)

  • bootmlx: bug fixes
  • getEstimatedIndividualParameters2: bug fix (eta’s were not correctly computed for normal distributions)

Rsmlx 1.1.0 – (on CRAN and GitHub, July 16th, 2018)

New functions:

  • whichPKmodel
  • readDatamlx
  • pkpopini
  • pkbuild

Rsmlx 1.0.2 – (on GitHub, July 6th, 2018)

  • bootmlx: bug fix: it is now possible to have correlations between random effects in the model.

Rsmlx 1.0.1 – (on CRAN and GitHub, June 22th, 2018)

Rsmlx 1.0.0 – (on GitHub, May 28th, 2018)