This PsN version has been tested on the following systems:

Windows XP 32-bit with ActivePerl 5.16.3, gfortran 4.6.0 and NONMEM 7.1.0, 7.1.2 and 7.2.0
Windows 7 64-bit with ActivePerl 5.20.1 (32 bit), gfortran 4.8.1 and NONMEM 7.3.0
Windows 7 64-bit with ActivePerl 5.16.3 (32 bit), gfortran 4.6.0 and NONMEM 7.3.0
Windows 7 64-bit with Strawberry Perl 5.18.2, gfortran 4.7.3 and NONMEM 7.2.0 and 7.3.0
MacOS X 10.10.3 with Perl 5.18.2, gfortran 4.8.0 and NONMEM 7.3.0
MacOS X 10.6.6 with ActiveState Perl 5.16.3, gfortran 4.6.2 and NONMEM 7.2.0 and 7.3.0
MacOS X 10.6.6 with ActiveState Perl 5.16.3, ifort 12.0.0 and NONMEM 7.2.0 and 7.3.0
Ubuntu Linux 14.10 with Perl 5.20.1, gfortran 4.9.1 and NONMEM 7.2.0 and 7.3.0
Fedora 20 with Perl 5.18.2, gfortran 4.8.2 and NONMEM 7.2.0
Scientific Linux 6.3 with Perl 5.10.1, gfortran 4.4.7 and NONMEM 7.2.0 and 7.3.0
CentOS 6.5 with Perl 5.10.1, gfortran 4.4.7 and NONMEM 7.1.0 and 7.3.0


************************
Installation
************************

Follow the installation instructions on
http://psn.sourceforge.net/install.php

************************
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 http://psn.sourceforge.net/

************************
Dependencies 
************************

- It is recommended to use Xpose version 4.5.3 for the -cwres functionality 
  and the default execute, vpc and bootstrap R plots.

************************
Known issues 
************************

- Windows users need to be careful to choose 
  a Perl version where the extra modules needed for PsN are available. See the 
  instructions on http://psn.sourceforge.net/install.php

- 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 4.4.0 to 4.4.8
*********************************************

   ___General___

- Bug fix: reading of data files which, excluding any header, contain a mix of comma,
  space and/or tabs, or multiple tabs, as separator between values. 
  Before the bug fix PsN would handle many sequences the same way as NONMEM, 
  for example <value><comma><space><value>, <value><comma><comma><value>, 
  <value><tab><value> and <value><space><space><value>, but not for example 
  <value><space><comma><value>, <value><tab><tab><value> or 
  <value><comma><tab><value>. 

  "Handle the same way as NONMEM" in this context means inserting the same 
  number of missing values in a particular sequence of separators. NONMEM 7.3
  will insert a missing value in <tab><tab>, <space><tab>, <comma><comma>, 
  <tab><comma> and <comma><tab>, but NONMEM will not insert a missing value
  in <tab><space>, <space><comma>, <comma><space> or <space><space>. 

  This bug can cause errors for programs that read the data file AND where the 
  unhandled combination of column separators appears before (to the left of) 
  the column(s) used by PsN. For example, execute is not affected because execute 
  does not parse the dataset. But bootstrap can be affected if option -stratify_on 
  is used AND the stratification column comes after the problematic separators,
  possibly leading to stratification being based on the wrong value. Scm can
  also be affected if the data filtering step is not run AND the unhandled combination
  appears before the covariate column(s), leading to the wrong min, max and median
  computed for that covariate.
  
- Bug fix: reading of $DATA when IGNORE=(list) has a list that is split over two or 
  more lines

- Bug fix: added file extensions .cpu, .agh and .shm to output files handled for NM7

- Bug fix: Correct search for nmfe script when major version is 6 and subversion is 
  defined in psn.conf.

- Bug fix: Relax checking of -nmfe_options so that -maxlim=(1,2) does not trigger warning.

- Bug fix: Reading of elapsed estimation time for SAEM.

- Bug fix: Reading of total run time when run time is 0 seconds.

- Bug fix: Parsing of $SIGMA/$OMEGA when comma is used instead of space as in $SIGMA 1,FIX

- Bug fix: Do not require that psn.conf exists during installation

- Bug fix: When reading matrices from lst-file, handle also NONMEM's sparse matrix format

- Bug fix: Reading of matrices from lst-files when .cov, .coi, .cor not available and 
  	  	   $OMEGA and/or $SIGMA has either at least one block of at least size 3, 
		   or two or more blocks of at least size 2.

- Bug fix: Parsing of ext-file from NM7.1.2 when there are multiple $PROB with one 
  	  	   $EST MAXEVAL=0 each, as in for example sir program when option 
		   -problems_per_file is > 1, and -dofv functionality of bootstrap. 

- Bug fix: Parsing of ext-file from NM7.2.0 and NM7.3.0 for $PROBLEMs where a $SIM 
  	  	   is followed by at least two $EST where the last $EST has MAXEVAL=0 and 
		   at least one previous in the same $PROBLEM does not have MAXEVAL=0.

   ___mcmp___

- Bug fix: allow option -significance_level=0.1 which did not work

- Add new significance levels 20, 15 and 10% to existing set of 5,1 and 0.1%. Updated 
  rplots template to include the new significance levels.

- Updated documentation.


   ___sumo___

- Bug fix: Check for high correlations. In older versions high correlations would not 
  be reported. 

- Bug fix: Check for large relative standard errors. In older versions many large 
  SEs were missed.

- Bug fix: Make sure omega/sigma correlations are computed also when variances are small. 


   ___vpc/npc___

- Bug fix: Computation of median of percent censored observations when using option 
  -censor.

- Bug fix: Do not warn that $INPUT uses synonym for DV when -dv is set on command-line.

- Bug fix: Handle option combination -bin_by_count=1 -bin_array=<> 

- Bug fix: Handle option -flip_comments in combination with $DATA that has file name 
  enclosed by quotes.


   ___scm___

- Bug fix: Use -threads set on command-line also in first iteration.

- Bug fix: In cases when data filtration step is run, handle $INPUT with items of the 
  form NAME=DROP or NAME=SKIP, in addition to just DROP or SKIP without NAME.

- Removed the possibility to set foce=0.

- Some clarifications in the documentation regarding -time_varying, documentation
  regarding METHOD=ZERO in combination with -linearize.

   ___boot_scm___

- Bug fix: Handle option -time_varying for regular covariates in boot_scm. 

- Bug fix: Not supplying valid_states caused boot_scm to crash. Adding same default values as for scm on valid_states. 

- Documentation for known bug that boot_scm does not handle dummy covariates which are time-varying.

- Removed the possibility to set foce=0.

   ___xv_scm___

- Removed the possibility to set foce=0.

   ___sse___

- New option -append_columns=col1,col2,... A comma-separated list of parameters to append 
  to the simulated data sets, if desired.

   ___update_inits___
	 
- Bug fix: Rounding of negative values and values with absolute value >= 10

- New option -flip_comments

- New option -nm_version, so that rounding of initial estimates is done based on major 
  NM-version (7 or less than 7)

   ___execute___
	 
- Bug fix: minor fix -rplots when detecting xpose run number from table file names: 
  do not include 'sim' in the run number, i.e. for sdtab65sim the run number is just 65

- Documentation that parentheses in -nmfe_options must be escaped on Unix/Linux.

   ___sir___
	 
- New feature: run multiple iterations where resampled parameters 
  from one iteration is used to generate proposal density for the next. 
  The resampled parameters are Box-Cox transformed to normality so that a 
  built-in Perl function for sampling from a multivariate normal can be 
  used in the next iteration. 

- Bug fix: Detection of NaN that string evaluate to 0 (on Windows).

   ___precond___
	 
- Some clarifications in the documentation.

   ___bootstrap___
	 
- Better documentation on the contents of bootstrap_results.csv

   ___runrecord___

- Bug fix: Correction in documentation so that tags for interindividual and interoccasion 
  variability in documentation match what is expected in the code

   ___nmoutput2so___

- Add DVID to Preditions and Residuals if present in sdtab

- Add support for multiple DVs in simulations

- Add CWRES if present in sdtab

- Residuals renamed from Residual
  
- Add bootstrap results file to RawResults

- Support for SO 0.2

- Added .ext as a rawresults file

- Added messages for all special flags and values from raw_results i.e. 
  covariance_step_warnings, s_matrix_singular etc

- Bugfix: RSEs are now a percentage

- Added bootstrap means and medians (SO 0.2)

- New option so_version to control the version of the so format to create

- Remove dosing rows for SimulatedProfiles and Residuals

- Added SimulationBlock/Dosing for plain AMT column

- Added SimulationBlock/PopulationParameters

- Option -precision removed. Now Using maximum precision available