Why are we doing this? – here is the story…

Dr. David W.A. Bourne (Department of Pharmaceutical Sciences, College of Pharmacy, The University of Oklahoma, Health Sciences Center) started it all.  The first time I used Boomer/Multiforte was about 1992 when I worked with Dr. Bourne for my Ph.D. program.  After graduation, I used Boomer when teaching a graduate PK/PD modeling course 1996.  There has been a very nice engine inside of either Boomer or MultiForte.   Back to when I was working with Dr. Bourne, I wondered if we could build GUI for Boomer working in MS-windows.  It would be certainly more friendly for users to use.  In 1992-3, I have been thinking using MS Visual BASIC to write the graphic user interface (GUI) for Boomer.  For some reasons, nothing was done until now.  However, it is always my dream to create a nice GUI for Boomer.  After 10 more years, we started designing and implementing the GUI for Boomer.  The other motivation to do this is because fewer and fewer users or students, including graduate students, now are familiar with MS-DOS or its commands.  They just know window’s graphic interface.  And when teaching PK/PD modeling with Boomer, I have to use its .BAT file explain model itself.  That’s quite unimaginable for students to understand. I usually have to spend a lot of time to explain how to operate DOS commands under Window to run Boomer.  The MS-DOS seems a long-time story that cannot motivate students’ interests any more.  Why should we choose Boomer as the teaching tool in PK/PD modeling?  It is because Boomer has been built-in many algorithms for numerical integration and nonlinear regressions (including Bayesian estimation algorithm) which are not available from other commercial packages. The other reasons includes that Boomer is accessible and is affordable (used to be shareware but full functions at that moment; and now it is a freeware).  Boomer is a cross-platform PK/PD modeling in Macintosh and PC with various OS.  Now as you may already find out that Boomer can run on Mac in both MacOS or MacOS X, and PC under command window of Win 9.x/ME/NT/XP/Linux (Fedora Core 2).  The only disadvantage may be that Boomer is DOS command line mode.  Thus we chose Java to build this GUI.  We call this GUI as JGuiB which stands for Java GUI for Boomer (or JGB).  When we started to build JGuiB, we had the first problem: Boomer cannot run with the command-mode with all its platforms!  That means one has to run Boomer first, then supply .BAT file or manually key in everything (including data) needed to complete the job.  We express what we tried to do to Dr. Bourne.  So Dr. Bourne decided to modify Boomer to make it possible to run with command-line (quiet) mode (see http://www.boomer.org for Boomer update history and other detailed information), and he completed the programming work just within few days.  The command-line mode made the dream come true.  Then Dr. Bourne worked on Linux Boomer for PC.  So we can have Linux Boomer now.  JGuiB may not have all fulfill all functions that Boomer has now, but we promise to make it better and better in the future.   JGuiB is also the main part of Mr. Lu, YH's MS thesis during 2003-2005 when he worked with me at College of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan.  Finally, we strongly suggest that you should visit Boomer website FIRST to view its online Manual.  That online manual is quite helpful and very well-organized.  -- edited Dec. 29, 2007.

    Why Using JGuiB with Boomer?

  1. include most commonly used functions of Boomer in PK/PD modeling included: normal fitting, simulation and Bayesian estimation;

  2. be much easier to check/examine your PK/PD models using self-explainable “Parameter Table” in JGuiB;

  3. be more friendly to cope with functionalities of Boomer;

  4. use less inputs than required: there is no need to assign “upper limit” or “lower limit” for each;

  5. do not need adjustable parameters.  JGuiB will take care of it for you.  In Bayesian estimation, JGuiB will even assign the initial values for your defined parameters if you don’t want to do this yourself;

  6. be much easier to handle your data sets (data I/O, copy & paste, etc..);

  7. handle more than one PK/PD models (max. 4 models) with the same data set using various weighting schemes in one project file can be defined;

  8. be more convenient in computer-aided teaching of PK/PD modeling; and

  9. be much more affordable than other commercial packages.