This PsN version has been tested on the following systems: Windows 10 64-bit with Strawberry Perl 5.22.3 , gfortran 4.9.2 and NONMEM 7.3.0 and 7.4.3 Windows 7 64-bit with Strawberry Perl 5.24.1, gfortran 4.9.2 and NONMEM 7.3.0 and 7.4.3 MacOS X 10.13.4 with Perl 5.18.2, gfortran 6.3.0 and NONMEM 7.4.3 Ubuntu Linux 17.04 with Perl 5.24.1, gfortran 6.3 and NONMEM 7.3.0 and 7.4.3 CentOS 7 with Perl 5.16.3, gfortran 5.1.0 and NONMEM 7.3.0 and 7.4.3 ************************ Installation ************************ Follow the installation instructions on https://uupharmacometrics.github.io/PsN/install.html ************************ Documentation ************************ All documentation is found in the doc subdirectory of the PsN installation directory. During installation it is also possible to have all documentation copied to a user-specified directory. All documentation can also be downloaded from the PsN website https://uupharmacometrics.github.io/PsN/docs.html ************************ Dependencies ************************ - It is recommended to use Xpose version 4.5.3 or later for the -cwres functionality and the default execute, vpc and bootstrap R plots. - Each userguide lists the R packages required to obtain the diagnostic R plots. ************************ Known issues ************************ - Windows users need to be careful to choose a Perl version where the extra modules needed for PsN are available. For new Perl installations the PsN team recommends Strawberry Perl See the instructions on https://uupharmacometrics.github.io/PsN/install.html - Installation problem on some Windows systems where perl 'system' command does not work. Make sure perl module File::Copy::Recursive is installed *before* (re)running the installation script setup.pl ********************************************* Changes and bug fixes from PsN 4.7.0 to 4.8.0 ********************************************* ___General___ - New required perl module YAML::XS - PsN will now give an error if an OMEGA or SIGMA BLOCK contains multiple FIX This would be inline with what NM-TRAN does. - PsN runs now generate meta.yaml to systematically describe various metadata. See the psn.pdf for documentation. - Fix bug causing datasets not to be filtered properly if ignore statement was written as IGN(VISI>3), i.e. with no equal sign. - The psn_installation.pdf has been removed and all information it contained has been merged into the installation section of the webpage. - The PsN_and_NONMEM7.pdf has been removed and all information it contained has been moved to PsN.pdf and other documentation. - All userguides now have a table of content - The vpctable script was removed from PsN - The data_stats script was removed from PsN - The -d2u common option was removed. PsN should support all encodings and dos2unix can be used separately if needed ___common_options___ - New option -model_subdir to use an extra directory level between the calling directory and the run directory. - The parafile option can now take non-trivial relative paths - When the parafile option is not used it will no longer be shown as 'none' in version_and_option_info.txt or meta.yaml ___NONMEM output parsing___ - Fix bug causing warnings when any parameter estimate was 'NaN' ___nonpb___ - Fix broken -lst_file option ___runrecord___ - No longer requires YAML::Tiny for csvy output. Will use the required YAML::XS ___frem___ - Postfrem now always creates its directory in the frem directory - The option -model to postfrem was removed. - The option -final_model was added to postfrem ___proseval___ - Bugfix: proseval would crash if no filtering was needed due to no accept or ignore in $DATA - Add a counter of the current number of observations to the abbreviated code. ___scm___ - Bugfix: Remove IGNORE or ACCEPT statements for linearized models that would else crash. - Linearized scm will always evaluate the derivatives model and not estimate it. - New option nointer to not use interaction for linearized scm - Linearized scm supports finding parameter-eta relation in two steps i.e.: ETA_CL = ETA(1), TVCL = THETA(1), CL = TVCL * EXP(ETA_CL) will properly be detected. Search is also done in $ERROR - Bugfix: Models with IGNORE=() and IGNORE=C where C is not @ caused scm to crash - Changed boundaries for covariate effect theta on exponential effect to be adaptive given an effect size within 0.01 and 100 This to reduce the risk of overflow. - Fix formatting of OFV values in scm log to avoid two numbers be put directly next to each other - Add support for categorical covariates with more levels than 2 for linearized scm ___linearize___ - Ignore all MDV!=0 in linear model if MDV obtainable - Support models with extension .ctl - Do not propagate name of -dir to subtools (i.e. underlying scm directory will always be called scm_dir1) - Only evaluate (MAXEVALS=0) the derivatives model. If lst file is available update initial values first. - New option nointer to not use interaction - Do not use COM in derivatives.mod to avoid name clash - Avoid NONMEM crash by adding dummy thetas for omegas that have not ETA(X) in code. ___nca___ - Set number of samples to default 300. ___runrecord___ - Fix bug causing est_time and cov_time to not be added to table for NONMEM 7.4 ___precond___ - Give error if model does not have any THETAS (i.e. linearized model) ___cdd___ - Fix critical bug (#50) casusing dofv to be calculated in a wrong way if any individual was ignored. - Do not propagate -dir to subtools (i.e. subdir of modelfit of cdd runs will always be 'modelfit_dir1') ___nmoutput2so___ - Bugfixes for simulation results that caused nmoutput2so to crash - Change columnType for TIME in SimulatedProfiles from time to idv ___transform___ - Added remove_0fix_omegas, tdist, full_block, add_tv, add_iov, iiv_on_ruv, power_on_ruv, remove_nonobs and omit_ids transformations ___resmod___ - Don't run dtbs and power for models giving negative IPRED ___qa___ - New tool