QMRF Documentation
Please find below
documentation of the Danish (Q)SAR Database (http://qsar.food.dtu.dk)
models for the endpoints included in the Danish (Q)SAR Models website (https://qsarmodels.food.dtu.dk). The
documentation is given in the QSAR Model Reporting Format (QMRF) and available
from the links in the Software column in the table. Please note that the Danish
(Q)SAR Models website uses versions of these models
implemented in Leadscope Predictive Data Miner / LSE v. 3.5.3-5 and based on
the same training sets and modeling approach. As reported in the QMRFs, the
cross-validation results were in most cases obtained using earlier versions of
Leadscope.
|
Endpoint |
N in training
set |
Software |
Cross validation result (%)a |
|
Not ready biodegradability (POS=Not Ready) |
735 |
Sens=87.3, Spec=85.2, Conc=86.4 |
|
|
Cytochrome P450 2D6
(CYP2D6) substrates
(human clinical data) |
746 |
Sens=60.0, Spec=89.4, Conc=80.1 |
|
|
Cytochrome P450 2C9 (CYP2C9) substrates
(human clinical data) |
736 |
Sens=30.0, Spec=89.6, Conc=75.4 |
|
|
Maximum recommended
daily dose (MRDD) in
humans ≤
2.69 mg/kg-2bw/d |
1222 |
|
Sens=78.6, Spec=82.5, Conc=80.7 |
|
Severe skin irritation in rabbit |
836 |
Sens=79.5, Spec=81.7, Conc=80.6 |
|
|
Respiratory sensitisation in humans |
80 |
Sens=91.7, Spec=95.5, Conc=93.9 |
|
|
Estrogen Receptor α binding (human
in vitro) ALL |
802 |
Sens=75.2, Spec=90.1, Conc=84.7 |
|
|
Estrogen Receptor α binding (human
in vitro) Balanced |
595 |
Sens=83.7, Spec=89.0, Conc=86.3 |
|
|
Estrogen Receptor α activation (human in vitro) |
481 |
Sens=73.1, Spec=86.6, Conc=80.7 |
|
|
Estrogen Receptor activation (in
vitro, CERAPP data) |
1420 |
Leadscope |
Sens=78.5, Spec=96.7,
BA=87.6 (2*5-fold cross-validation) |
|
Androgen Receptor
inhibition (human in
vitro) |
874 |
Sens=51.7, Spec=91.2, Conc=80.4 |
|
|
Androgen Receptor
binding (in vitro, CoMPARA data) |
1662 |
Leadscope |
Sens=80.4, Spec=93.3, BA=86.9 (2*5-fold cross-validation) |
|
Androgen Receptor
activation (in vitro, CoMPARA data) |
1659 |
Leadscope |
Sens=76.3, Spec=99.3, BA=87.8 (2*5-fold cross-validation) |
|
Androgen Receptor
inhibition (in vitro, CoMPARA data) |
1525 |
Leadscope |
Sens=81.5, Spec=92.2, BA=86.9 (2*5-fold cross-validation) |
|
Arylhydrocarbon (AhR)
Activation – Rational final model |
4625 |
Leadscope |
Sens=75.4, Spec=87.0, BA=80.8 (2*5-fold cross-validation) |
|
Arylhydrocarbon (AhR)
Activation – Random final model |
4625 |
Leadscope |
Sens=86.4, Spec=91.5, BA=88.9 (2*5-fold cross-validation) |
|
Thyroperoxidase (TPO)
inhibition QSAR1 (rat in vitro) |
877 |
Sens=72.4, Spec=89.0, BA=80.6 Ext. validation: Sens=79.7,
Spec=90.8, BA=85.3 |
|
|
Thyroperoxidase (TPO)
inhibition QSAR2 (rat in vitro) |
1519 |
Sens=75.6, Spec=89.8, BA=82.7 |
|
|
Pregnane X
receptor binding (human
in vitro) |
631 |
Sens=80.4, Spec=80.4, Conc=80.4 |
|
|
Pregnane X
Receptor (PXR) Binding (Human in vitro) NEW |
1504 |
Sens=86.6, Spec=98.5, Conc=92.6 (2*5-fold cross-validation) |
|
|
Pregnane X
Receptor (PXR) Activation (Human in vitro) |
2176 |
Sens=89.1, Spec=98.6, BA=93.9 (2*5-fold cross-validation) |
|
|
Pregnane X
Receptor (PXR) Activation (Rat in vitro) |
2330 |
Sens=86.5, Spec=97.4, BA=92.0 (2*5-fold cross-validation) |
|
|
CYP3A4
Induction (Human in vitro) |
2271 |
Sens=86.7, Spec=98.2, BA=92.5 (2*5-fold cross-validation) |
|
|
Constitutive Androstane Receptor (CAR) activation at max. 20 µM |
924 |
Leadscope |
Sens=72.2, Spec=93.5, BA=82.8 (2*5-fold cross-validation) |
|
Constitutive Androstane Receptor (CAR) activation at max. 50 µM |
1903 |
Leadscope |
Sens=78.4, Spec=91.4, BA=84.9 (2*5-fold cross-validation) |
|
Constitutive Androstane Receptor (CAR) inhibition at max. 20 µM |
1408 |
Leadscope |
Sens=58.4, Spec=97.1, BA=77.8 (2*5-fold cross-validation) |
|
Constitutive Androstane Receptor (CAR) inhibition at
max. 50 µM |
1870 |
Leadscope |
Sens=72.4, Spec=91.6, BA=82.0 (2*5-fold cross-validation) |
|
Bacterial reverse mutation test (Ames test in S. typhimurium in vitro) |
4102 |
Sens=84.3, Spec=85.7, Conc=84.9 |
|
|
Chromosome
aberrations in CHL
cells (in
vitro) |
600 |
Sens=74.6, Spec=75.2, Conc=74.9 |
|
|
Mutations in thymidine kinase locus in mouse
lymphoma cells (in vitro) |
555 |
Sens=85.1, Spec=83.8, Conc=84.4 |
|
|
Mutations in HGPRT locus in CHO cells
(in vitro) |
239 |
Sens=81.7, Spec=78.4, Conc=80.5 |
|
|
Unscheduled DNA synthesis (UDS) in rat hepatocytes (in vitro) |
415 |
Sens=74.1, Spec=70.1, Conc=72.4 |
|
|
Syrian hamster embryo
(SHE) cell transformation (in vitro) |
363 |
Sens=71.6, Spec=76.5, Conc=74.5 |
|
|
Sex-linked recessive lethal
(SLRL) test in Drosophila m. (in vivo) |
367 |
Sens=79.1, Spec=80.3, Conc=79.6 |
|
|
Micronucleus test in
mouse erythrocytes (in vivo) |
357 |
Sens=64.1, Spec=77.6, Conc=72.3 |
|
|
Dominant lethal
mutations in
rodents (in
vivo) |
191 |
Sens=61.5, Spec=80.4, Conc=71.8 |
|
|
Sister chromatid exchange in
mouse bone marrow cells (in
vivo) |
265 |
Sens=88.6, Spec=95.9, Conc=94.0 |
|
|
Comet assay in mouse (in
vivo) |
286 |
Sens=86.6, Spec=80.8, Conc=83.1 |
|
|
Liver specific cancer in rat
or
mouse (in vivo) |
320 |
Sens=35.6, Spec=88.6, Conc=69.3 |
a Sens: sensitivity; Spec: specificity; Conc: concordance.
Unless otherwise stated, 5 * 2-fold cross-validation was performed.