58
S. A. Khan et al.
Abbreviations
ARD
Automatic Relevance Determination
CCLE Cancer Cell Line Encyclopedia
CMap Connectivity Map
CP
Canonical decomposition and Parallel factor analysis
DILI
Drug-induced liver injury
FA
Factor Analysis
FCFP Functional Connectivity Fingerprints
FDA
Food and Drug Administration
GFA
Group Factor Analysis
GI50
50% Growth Inhibition
IPA
Ingenuity Pathway Analysis
LC50 50% Lethal Concentration
LDA
Latent Dirichlet Allocation
LINCS Library of Integrated Network-Based Cellular Signatures
MF
Matrix Factorization
MoA
Mode of action
MTF
Multi-tensor Factorization
NCI
National Cancer Institute
PCA
Principal Component Analysis
PTGS Predictive Toxicogenomics Space
QSAR Quantitative Structure–Activity Relationship
TF
Tensor Factorization
TGI
Total Growth Inhibition
4.1 Introduction
Cellular responses to drugs and other chemical compounds are increasingly being
measured at multiple levels of detail and resolution. For instance, ex vivo toxicity
measurements summarize the phenotypic responses in human primary cells [1, 2],
while profiling of genome-wide transcriptomic responses opens up a system-level
view to the compounds’ mode-of-action (MoA) mechanisms. The study of relationships between genome-wide genomic or molecular responses of the cells to exposure
to substances and the corresponding toxicological outcomes is referred to as toxicogenomics. Understanding these complex relationships can not only identify the
molecular mechanisms behind toxicity but also suggest ways to avoid toxic effects
in medical or other applications [3–7]. Toxicogenomics may be especially pertinent
for analyzing data from cellular assays, and for reducing and eventually replacing
the use of animal experiments for toxicity testing during drug development, also
referred to as 3R approaches [3, 4, 6]. The reductions in the costs of genomics and
transcriptomic assays are enabling factors toward 3R as well [6, 8].
S. A. Khan et al.
Abbreviations
ARD
Automatic Relevance Determination
CCLE Cancer Cell Line Encyclopedia
CMap Connectivity Map
CP
Canonical decomposition and Parallel factor analysis
DILI
Drug-induced liver injury
FA
Factor Analysis
FCFP Functional Connectivity Fingerprints
FDA
Food and Drug Administration
GFA
Group Factor Analysis
GI50
50% Growth Inhibition
IPA
Ingenuity Pathway Analysis
LC50 50% Lethal Concentration
LDA
Latent Dirichlet Allocation
LINCS Library of Integrated Network-Based Cellular Signatures
MF
Matrix Factorization
MoA
Mode of action
MTF
Multi-tensor Factorization
NCI
National Cancer Institute
PCA
Principal Component Analysis
PTGS Predictive Toxicogenomics Space
QSAR Quantitative Structure–Activity Relationship
TF
Tensor Factorization
TGI
Total Growth Inhibition
4.1 Introduction
Cellular responses to drugs and other chemical compounds are increasingly being
measured at multiple levels of detail and resolution. For instance, ex vivo toxicity
measurements summarize the phenotypic responses in human primary cells [1, 2],
while profiling of genome-wide transcriptomic responses opens up a system-level
view to the compounds’ mode-of-action (MoA) mechanisms. The study of relationships between genome-wide genomic or molecular responses of the cells to exposure
to substances and the corresponding toxicological outcomes is referred to as toxicogenomics. Understanding these complex relationships can not only identify the
molecular mechanisms behind toxicity but also suggest ways to avoid toxic effects
in medical or other applications [3–7]. Toxicogenomics may be especially pertinent
for analyzing data from cellular assays, and for reducing and eventually replacing
the use of animal experiments for toxicity testing during drug development, also
referred to as 3R approaches [3, 4, 6]. The reductions in the costs of genomics and
transcriptomic assays are enabling factors toward 3R as well [6, 8].