| Question |
Answer |
| Can I get a copy of the model files you used in the
seminar? |
Click to download the model
files. |
| Can I get a copy of the slides you used in the seminar? |
Click to download the seminar
slides. |
| I would like to view a recorded copy of the webinar. |
Click here to view |
| Can I get a demo version of acslX? |
For a free software trial, click
here. |
| Where can I get pricing information? |
Please contact a sales represeative to
recieve a quote to meet your exact software configuration needs. |
Do you support population modeling?
|
We're currently finishing a population modeling tool which will
initially support nonparameteric modeling using Bayesian techniques
and MCMC sampling. The tool will also support parallel (cluster computer
based) execution of models. Models will be defined using a langauge
created for the specification of hierarchical statistical models.
Pilot users will be receiving the first version of the tool within
the next two months, and the tool should be widely available by the
end of the year. |
| Where can I find example applications? |
Ask us for an evaluation version of acslX, and have a look at the
example applications that are installed with the program. Also try
using your favorite simulation engine to search the web for ACSL
models; you'll find many example there as well. We also have a library
of technical publications on out website, many of which include discussions
on models and source code listings. |
| Does your software integrate with any 3rd party products? |
We currently have interfaces within acslX to MS Excel and MS Access,
using special M functions. For MS Access, M functions are avaialble
to execute SQL statements to either retrieve data from a database
into an M variable, or to store data contained in a M variable into
the databased. The MS Excel interface functions similarly allow data
in M variables to be stored to a range of cells in an Excel spreadsheet,
or retrieved from a range of cells. We also have an Excel plugin
which allows acslX simulations to be controlled and executed from
Excel spreadsheets. Cells in the spreadsheet are used to set the
values of constants in the acslX simulation; similarly, values and
time histories of model outputs are returned to Excel and placed
in spreadsheet cells. This allows the complete mathematical and analysis
capabilities of Excel to be applied to acslX simulation data. |
| During Parameter optimization, when you do different
doses, do you want seperate files for parameters that you validate
to like you have for the different doses |
Yes, for the PE wizard youll need to separate the data into different
files to apply the descriptor value for each data set. |
| What operating systems do you support? |
The full version show during the webinar runs on various versions
of MS Windows. We also support a command-line version of acslX which
includes the interpreter, translator and simulation runtime, and
which can be executed from a command prompt on many Linux distributions.
This version uses GNU/Linux tools for things such as source file
editing and plotting, and can be used as a batch-mode application
for excution on Linux-based cluster computers. |
| What functionality is included with acslX, Optimum
and PK toolkit? |
acslX is the basic software package. It contains the environment
to build, execute and analyze models.
acslX Optimum is an enhancement to the comprehensive analysis
tools provided in acslX. These include robust parameter
estimation, min/max optimization, sensitivity analysis, and Monte
Carlo analysis.
The PK/PD toolkit is a set of PowerBlocks commonly used in PK/PBPK/PD
modeling.
|
When we use descriptors in parameter estimation wizard,
is the model putting different values for that descriptors?
|
Assuming you have two CSL constants for IVDOSE and ORAL dose
and corresponding data sets for each pathway exclusively the way
I would first recommend handling the descriptors for it would be
like this:
|
| Is Matlab a requirement for acslX. Are the scripts
between them compatable? |
acslX has it's own analysis environment and doesn't require
Matlab. The scripting language that acslX uses is closely related
to the Matlab language.
acslX models can be imbedded into Simulink. We are working
on an interface wizard that would simplify this process.
|
| Is there a limit on the number of parameters estimated
as in older ACSL versions? |
There isn’t a set limit on the number of parameters to be
estimated simultaneously – however as you estimate more parameters
simultaneously the number of simulation runs required to complete
the optimization increases dramatically (depending on the algorithm
you’ve selected). |
Is the regression weight (called here "Heteroscedasticity
parameter") estimated as well if you uncheck the respective
box?
|
You can include the Heteroscedasticity Parameter to the list of
parameters the optimization algorithm uses to reduce the LLF, and
you are correct when you guessed that is done by checking the ‘Vary
Heteroscedasticity Parameter’ checkbox. |
| Can you simulate from a multivariate normal-distribution? |
yes, we support the multivariate normal distribution, but to construct
a Monte Carlo analysis which uses it you have to specify the analysis
using an M script (as opposed to letting the wizard generate it).
Likewise, custom distributions, empirical distributions or other
correlations between input parameters can be prescribed by specifying
the Monte Carlo analysis in an M script. |
| Was the "pe300.m" script used for the Optimum
procedure automatically generated or was it created separately? |
The pe300.m file was automatically generated by the Parameter Estimation
Wizard for this case. It is just m-code though, so you could use
that file as a template and modifiy it to do some other optimization. |
| Where are the optimum files that are generated by the
PE wizard saved? How are the files named? |
Those m-files are created and saved in the project folder (the
folder where the corresponding acslX project is) when the wizard
is finished. The filename is the same as the one given when the PE
wizard is launched. |
| Can constants be specified prior to the dynamic section? |
Constants can still be specified anywhere in the code- the translator
will strip them out and put them into memory early on in the build
process. |
