[NMusers] PLT Data - A Graphical Interface to Create NONMEM Datasets (beta version)

From: Dennis Fisher <fisher_at_plessthan.com>
Date: Mon, 22 Jun 2015 06:55:36 -0700

Fellow NONMEM users,

NONMEM users are well aware that creation of a NONMEM dataset can be a =
complicated task, often requiring many hours or days. PLT Data is a =
graphical interface designed to facilitate that task. Source data can =
be in a variety of formats (SAS7BDAT, XPT, CSV, TXT, tab-delimited, =
pipe-delimited). PLT Data reads these files, saving all output in CSV =
format and provides the user with information about the content of each =
file. The user then populates fields in the interface, directing PLT =
Data how to process the dataset. For example, the user identifies which =
file contains Cp data, column headers for each of date/time, =
concentration, etc., and how date/time are formatted. Similar steps are =
applied to the file containing dosing data and, if available, =
demographics, vital signs, and laboratory values. PLT data can even =
link files, one of which contains concentrations, the other containing =
sample times. PLT Data then constructs a dataset, formatted for NONMEM, =
containing the core information that the user expects and creates a =
series of graphics (e.g., by subject, by dose group, composite, =
time-after-dose).

PLT Data also calculates time-after-dose, number of doses, dose #, and a =
variety of covariates such as GFR (using several user-selectable =
formulae), "elderly", lean body mass, and BMI. If LOQ is provided, PLT =
Data can create an EVID column based on Beal's Method 5 or 6. PLT Data =
also attempts to translate covariates in text form into numerics, e.g., =
if race appears in the source data as "White", "black", Caucasian", "W", =
or "B", PLT Data attempts to map these to numbers (and informs the user =
as to how that was accomplished). PLT Data can add records to the =
dataset with EVID=2; these records provide additional predictions, =
thereby allowing the display of graphics better representing the =
predicted Cp profile. This is but a small subset of the many things =
that can be included in a dataset. PLT Data also summarizes the data =
(number of doses / subject; total dose / subject; observations / =
subject) and each categorical or continuous covariate. In addition, PLT =
Data provides a record of every step that it executes, thereby ensuring =
traceability and reproducibility.

The engine for PLT Data is R (open-source software available at =
R-project.org). Other than installation of R, under most circumstances, =
the user does not need to be facile with R. However, in some instances, =
construction of more complicated datasets requires that the user write a =
few lines of R code (examples are provided).

PLT Data is in the final stages of beta testing, having been tested on > =
30 datasets. It is available for free at www.PLTsoft.com; an installer, =
examples, and a manual are provided. Details of what can be =
accomplished with PLT Data are available at PLTsoft.com.

I look forward to people using PLT Data. If you encounter problems (as =
one expects in the final stages of beta testing), I will fix them. And, =
I encourage feedback from users as to how to improve the interface and =
code.

Dennis



Dennis Fisher MD
P < (The "P Less Than" Company)
Phone: 1-866-PLessThan (1-866-753-7784)
Fax: 1-866-PLessThan (1-866-753-7784)
www.PLessThan.com <http://www.plessthan.com/>




Received on Mon Jun 22 2015 - 09:55:36 EDT

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:44:50 EDT