Work Package 8: Documentation

/ Project: / DDMoRe
Work Package: / Work package 8
Document name: / Glossary
Document version: / 1.1

Glossary

Work Package 8: Documentation

Author: / Jonathan Chard
Prepared by: / Wendy Aartsen, Eva Potters
Approved by:

Table of contents

1Definitions

1.1Acronyms and Abbreviations

1.2Definitions and System Names

2Document Revision History

1Definitions

1.1Acronyms and Abbreviations

Acronym / Definition
21 CFR Part 11 / 21 Code of Federal Regulations (21 CFR Part 11) Electronic Records; Electronic Signatures
AE / Adverse Event
AIC / Akaike Information Criterion
AMT / Amount of dose at a specified time
AoI / Area of Interest
APE / Absolute Prediction Error
API / Application Programmers Interface
Interface exposed to a programming language to allow tools to be accessible to other software
ASCII / American Standard Code for Information Interchange
BBB / Blood Brain Barrier
BIC / Bayesian Information Criterion
BQL / Below quantifiable limit
BURNIN / Number of MCMC iterations to discard before sampling for inference (MCMC method)
CI / Confidence Interval
CLI / Command Line Interface
CNS / Central Nervous System
Connector / A piece of software which enables modelling software to communicate with the interoperability framework
CRD / Completely Random Dropout
CRM / Continuous Reassessment Method
CSDISC / Clinical Data Interchange Standards Consortium
CSV / i] Comma separated values
ii] Computerized system validation
CTS / Clinical Trial Simulation
DAE / Differential-Algebraic Equation
DDE / Delay Differential Equation
DDMoRe / Drug Disease Model Resources
DMR / Data and Modelling Resource
Artefacts used throughout the modelling process, such as datasets, model definitions, and results.
DMX / Drug Model Explorer (see Names)
D-opt/G-opt / D-optimality / G-optimality (optimal design metric)
DoW / Description of Work
DV / Dependent Variable
EFPIA / European Federation of Pharmaceutical Industries and Associations
EMEA / European Medicines Agency
F2F / Face-to-Face
FDA / Food and Drug Administration
FIM / Fisher Information Matrix (optimal design metric)
FO / First Order
FOCE / First Order (Conditional Estimation)
FOCEI / First Order (Conditional Estimation) with Interaction
GLM / General Linear Model
GUI / Graphical User Interface
IDV / Independent Variable
IMI / Innovative Medicines Initiative
IMI-JU / Innovative Medicines Initiative Joint Undertaking
IQ / Installation Qualification (see Definitions)
ITER / Number of iterations (MCMC or NONMEM control)
JVM / Java Virtual Machine
LAN / Local Area Network
LLOQ / Lower Limit of Quantification
M&S / Modelling and Simulation
MAXEVAL / Maximum Evaluations
MCL / Model Coding Language
The MCL is a declarative language used to define the mathematical structure of the model
MCMC / Markov Chain Monte Carlo
MCPEM / Monte Carlo Parametric Expectation Maximization
MDL / Modelling Definition Language
The modelling definition is comprised of two parts, the MCL and the TEL.
MDV / Missing dependent variable
MML / Modelling Markup Language
The Modelling Markup Language is a machine-readable representation of the data and modelling resources created and stored in a project
MSE / Mean Squared Error
MTD / Maximum Tolerated Dose
NAS / Network Attached Storage
NDLM / Normal Dynamic Linear Model
NFS / Network File System
NLMEM /

A new SAS/IML macro for hierarchical nonlinear models.

NONMEM / NONLinear Mixed Effects Modelling (see Names)
NPC / Numerical Predictive Check
ODE / Ordinary Differential Equation
OFV / Objective Function Value (NONMEM output)
OPTDES / Optimal Design
OQ / Operational Qualification (see Definitions)
PBPK / Physiologically based pharmacokinetic(s)
PD / Pharmacodynamic(s)
PDE / Partial Differential Equation
PDF / Portable Document Format
PE / Prediction Error
PFIM / Population Fisher Information Matrix
PI / Prediction Interval
PK / Pharmacokinetic(s)
PKBUGS / Bayesian Inference Using Gibbs Sampling (PharmacoKinetics) (see Names)
PKS / Pharsight Knowledgebase Server (see Names)
PPC / Posterior Predictive Check
PQ / Performance Qualification (see Definitions)
PsN / Perl Speaks NONMEM (modelling software for managing NONMEM runs)
QQ (plot) / Quantile-Quantile plot
R / R (see Names)
RCP / Rich Client Platform
RD / Random Dropout
RSE / Relative Standard Error
RTF / Rich Text Format
An open document format supported by Word
SAEM / Stochastic Approximation Expectation Maximization
SAS / Statistical Analysis System (see Names)
SB / System biology
scp / Secure Copy
SD or Stdev / Standard Deviation
SDE / Stochastic Differential Equation
SDTM / Study Data Tabulation Model
SE / Standard Error
sftp / SSH File Transfer Protocol
SGE / Sun Grid Engine
SIG / Significant digits (NONMEM control statement)
SLURM / Simple Linux Utility for Resource Management
SMP / Symmetric Multiprocessing
SSH / Secure Shell
SPE / Standardized Prediction Error
SWT / Standard Widget Toolkit
TEL / Task Execution Language
The TEL is an imperative language that describes a sequence of task that perform some actions on the model described with the MCL
TC / Teleconference
VPC / Visual Predictive Check
WAN / Wide Area Network
WP / Work Package
WSDL / Web Service Definition Language
XML / eXtensible Markup Language
- (dash) / Indicates lack of response/data

1.2Definitions and System Names

System Name / Description
21 CFR Part 11 / FDA rule that defines parameters by which pharmaceutical companies can author, approve, store, and distribute records electronically.

Actor / In the context of software requirements analysis this is the role adopted by a person or system wheninteracting with a software system.
Adaptive
Random Search / Adaptive Random Search is a search technique in which the parameter space is investigated by drawing from a vector with large variances when far away from the global optimum, and when in the region of the global optimum, to reduce the variance by giving more detailed investigation.
Administrator / A specific user role with privileges to manage users, job queues, system access, etc.
Analysis
Dataset / A dataset (derived from an Import Dataset), which can be used for model estimation.
Bayesian
Inference / Bayesian Inference is statistical inference in which prior evidence or observations are used to update or to newly infer the probability that a hypothesis may be true.
Berkeley Madonna / Berkeley Madonna is a general-purpose differential equation solver and real-time simulation tool.

CDISC / A suite of global, platform independent data standards that enable information system interoperability to improve medical research and related areas of healthcare.
Central Integrated Database / A database within the system that holds all relevant information (Import Datasets, Analysis Datasets, model definitions, model output, scripts, etc).
Citrix / Citrix A suite of remote application delivery software developed by Citrix.

Comma- Separated Values (CSV) / Comma-Separated Values (also known as a comma-separated list) is a file format that stores tabular data as ASCII separated by commas.
Computerized System Validation
(CSV2) / The documented process of assuring that a computerized system does exactly what it is designed to do in a consistent and reproducible manner.
Data / The termdatarefers to qualitative or quantitative attributes of avariableor set of variables. Data (plural of "datum") are typically the results of measurements and can be the basis ofgraphs,images, or observations of a set of variables.
Data are often viewed as the lowest level ofabstractionfrom which information and then knowledge are derived.
Raw data, i.e. unprocessed data, refers to a collection ofnumbers,characters, images or other outputs from devices that collect information to convert physical quantities into symbols.
Data Views / Different formats for viewing data (e.g. spreadsheet format, etc).
Dataset / Adata set(ordataset) is a collection ofdata, usually presented intabularform. Eachcolumnrepresents a particular variable.
Eachrowcorresponds to a given member of the data set in question. Its values for each of the variables, such as height and weight of an object or values ofrandom numbers. Each value is known as adatum.
The data set may comprise data for one or more members, corresponding to the number of rows.
Design / The general framework for a randomized controlled experiment used to investigate a particular drug effect, including exposure-disease relationships.
Distributed
Computing / The distribution of computational workload between several computers in a cluster or grid.
DMX / Drug Model Explorer –a program for exploring drug characteristics based on modelling and simulation results.

D-optimality / Optimality criterion based on the maximization of the determinant of the Fisher information matrix (max det F), or, equivalently, minimizing the determinant of its inverse.
Eclipse RCP / A platform for building and deploying rich client applications.
ED-optimality / Optimality criterion based on the maximization of the expected value (over a given parameter distribution) of the determinant of Fisher information matrix.
Electronic
Signatures / Food and Drug Administration 21 CFR Sec. 11.3 definitions: Electronic signature means a computer data compilation of any symbol or series of symbols executed, adopted, or authorized by an individual to be the legally binding equivalent of the individual’s handwritten signature.
m?fr=11.100
End-User / A specific user role with privileges to manage only his/her own jobs queues, etc.
Estimation / A mathematical or statistical procedure for the derivation of model parameters, given a specific set of data.
Extensible Markup Language (XML) / An open standard for exchanging structured documents and data over the Internet that was introduced by the World Wide Web Consortium (W3C).
External
Authentication / The process of using the operating system to authenticate users to an application. External authentication eliminates the need for users to log in to both the operating system and the application.
Feature / A description of some functionality that will be part of the implemented system.
Federov
Exchange / This is an iterative algorithm that replaces at each iteration a design point with the point from a list of candidate points that improve the design most. To increase the probability of finding the global optimum, several tries are made using different starting designs.
Git / A distributed version control system.
Hyper- threading / Hyper-threading technology allows a single processor to run two threads at the same time.
Import Data File
Structure / A common structure of data files that can be imported into the system.
Import Dataset / A set of raw data imported into the system.
Input Data File
Types / Types of files that can be used when importing data.
Input data file types must include (but need not be limited to) the following: ASCII, Excel, SAS, S-plus, NONMEM (data file), SDTM (CDISC), PKS.
Installation Qualification (IQ) / A product setup qualification which checks that the correct items have been installed, that the site requirements for the system have been met, and that the system is functional.
Job Scheduling
Management / Functionality to manage job queues, and at any time prioritise, kill, suspend, and restart jobs.
Kerberos / Kerberos is the name of a computer network authentication protocol, which allows individuals communicating over an insecure network to prove their identity to one another in a secure manner, and also a suite of free software published by Massachusetts Institute of Technology (MIT) which implements this protocol.
LaTeX / A high quality typesetting system that includes features designed for the production of technical and scientific documentation.
MATLAB / MATLAB is a commercial “Matrix Laboratory”, which operates as an interactive programming environment.

Metadata / Metadata (metacontent) is defined as data providing information about one or more aspects of the data, such as:

• Means of creation of the data
• Purpose of the data
• Time and date of creation
• Creator or author of data
• Placement on acomputer networkwhere the data was created
• Standardsused

Methods of
Estimation / Numerical algorithms for solving nonlinear mixed-effects problems.
The Methods of Estimation supported by the system must include: FO, FOI, FOCE, FOCEI, LAPLACIAN, Lindstrom and Bates, MCMC (based methods), MCPEM, SAEM, Gaussian Quadrature, and nonparametric methods.
Model / A mathematical description of the behaviour of a system.
Module / An independent component of the system designed to facilitate a specific task.
Monte Carlo
Simulation / An analytical technique in which a large number of simulations are run using random quantities for uncertain variables, and looking at the distribution of results to infer which values are most likely.
Multitasking / The ability to run more than one task or program at the same time. In the case of this application it refers to batch jobs (causing no application locks) and multiple user sessions (concurrently on one desktop) to modify and run jobs at the same time.
NM-TRAN / The language used to define models that are executed with NONMEM.
NONMEM / NONMEM is the gold standard software for nonlinear mixed-effects modelling.

Numerical
Precision / Property of a numerical algorithm describing the precision with which results can be produced.
Numerical
Stability / Property of a numerical algorithm describing its ability to run without causing exceptions, its ability to converge, and reproducibility of its results.
Operational Qualification (OQ) / Tests to determine that the system performs against a specification in terms of functional tests performed on individual components of the system.
Oracle / Oracle Corporation is one of the major companies developing database management systems (DBMS).

Other Diseases area / Therapeutic areas different from diabetes and oncology
Output Data
File Types / Types of files that can be used when exporting data.
Output data file types must include (but need not be limited to) the following: ASCII, Excel, SAS, S-plus, NONMEM (data file), SDTM (CDISC), Word (for reports), MATLAB, R, PKS.
Performance Qualification (PQ) / Tests to verify that the systems and equipment, as connected together in the target environment, can perform effectively and reproducibly.
Pirana / Work on a cluster, or locally. Use the PsN-toolkit for advanced M&S. Uses R and Xpose to create plots

PKBUGS / Add-on package for WinBUGS dealing with the analysis of pharmacokinetic data.

PKS / Pharsight Knowledgebase Server – is a database system for managing PK/PD data.

PopDes / A program for Optimal Design of univariate and multivariate, Individual and Population Pharmacokinetic and Pharmacodynamic Experiment running under Matlab.
mailto:
Power / The power of a statistical test is the probability that the test will reject a false null hypothesis, or in other words that it will not make a Type II error. As power increases, the chances of a Type II error decrease, and vice versa.
Project / A project is classed as a single piece of work in which a user has a specific “question to answer” or a hypothesis to test.
R / R is a tool for statistical analysis and generation of graphs.

Relational Database Technologies / Database technology based on the relational model for database management, which is a data model based on predicate logic and set theory.
Result Set / A result set is generated by the execution of a task specified in the task execution language. A result set could be:
-A dataset (generated by some manipulation of an existing dataset, or generated using some simulation)
-A parameter estimation
-A model simulation
-Post-analysis on the outputs of a parameter estimation
Reviewer / A specific user role with read-only privileges.
Risk Analysis / A qualitative technique which involves a disciplined analysis of the events which could transform a potential hazard into undesired outcomes.
Sample Size / Sample Size, usually designated ”n,” is the number of individuals or repeated measurements in an experiment.
SAS / SAS is a comprehensive system for performing e.g. statistical analysis.

SDTM
Simplex / The simplex method (or Nelder-Mead or downhill simplex method) is a commonly- used nonlinear optimization algorithm for minimizing an objective function in a
multi-dimensional space.
Simulated
Annealing / An algorithm for solving “hard” problems (insoluble through numerical methods), notably combinatorial optimization, based on the metaphor of annealing.
Simulation / A mathematical or statistical procedure for the generation of hypothetical data given a specific model and study design.
S-PLUS / S-PLUS is a general purpose statistics package using the S programming language.

Step / A step in a project is composed of a set of working files, in particular:
-Zero, one or many datasets
-A model written in the MCL
-Zero, one or many tasks written in the TEL
-Zero, one or many result sets generated as a result of the execution of a task written in the TEL
Summary Level
Predictions / Synthesis of a simulation outcome, presented graphically or numerically.
Symmetric Multiprocessing (SMP) / The distribution of computational workload between separate processors that share the same physical memory.
System / A group, set, or aggregate of things, natural or artificial, forming a connected or complex whole.
System
Stability / The ability of a computer system to function appropriately with a given (high) workload, e.g. with a given number of concurrent users running a given number of simultaneous jobs.
Task
Term / Description
TRELLIS / TRELLIS is a powerful graphical toolkit used to display relationships in large datasets compactly within a single display, using panels to distinguish between different groups.
Type I Error / Also known as an "error of the first kind", an α error, or a "false positive": the error of rejecting a null hypothesis when it is true.
Type II Error / Also known as an "error of the second kind", a β error, or a "false negative": the error of not rejecting a null hypothesis when the alternative hypothesis is true.
Use case / A set of scenarios describing how the System will respond its environment; typically user interactionor interactions with other software systems.
User Roles / The different roles that users may undertake, e.g. administrator/end-user/reviewer. Each user role has different privileges in terms of the User Tasks he/she isallowed to perform.
User Tasks / The actions possible within the system, e.g. in terms of management of job queues.
Variance- covariance Matrix / The Variance-covariance Matrix is a matrix of covariances between elements of a vector of model parameter estimates given a model and design, the diagonals of which are variances.
WinBUGS / Windows implementation of the BUGS (Bayesian Inference Using Gibbs Sampling) project, concerned with flexible software for the Bayesian analysis of complex statistical models using Markov chain Monte Carlo (MCMC) methods.

Windows Active
Directory
Server / Active Directory is an implementation of LDAP directory services by Microsoft for use in Windows environments.
WinPOPT / WinPOPT and POPT are a set of programs written in MATLAB that can be used for optimization of nonlinear mixed effects models.

Workflow / The tasks, procedural steps, and organization, given the required input and output information, needed for each step of a data analysis process.
Xpose / Xpose is an R-based model building aid for population analysis using NONMEM.

2Document Revision History

Version / Author / Date / Description
0.1 / J Chard / 22/8/2011 / Initial version
0.2 / J Chard / 30/9/2011 / Incorporated glossary items added to the prototype and requirements document
0.3 / W. Aartsen / 23/12/2011 / Incorporated glossary items from MCL specification
0.4 / W. Aartsen / 16/07/2012 / Incorporated glossary items from MCL and TEL specification draft 3, documentation on the API
0.5 / W. Aartsen / 17/12/2012 / Incorporated glossary items from D8.59 Documentation of existing tools
Filename: d8.6 DDMoReGlossary_V1.1_WAA[1].doc / Date: / 1MAR2013

Page: 1 of 16