2,4,8,10-tetraoxa-3λ6,9λ6-dithiaspiro[5.5]undecane 3,3,9,9-tetraoxide

Density

Link to relevant study record(s)

Reference

Endpoint:

density, other

Type of information:

(Q)SAR

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

results derived from a valid (Q)SAR model and falling into its applicability domain, with adequate and reliable documentation / justification

Justification for type of information:

moderate reliable QSAR study available

Principles of method if other than guideline:

QSAR

GLP compliance:

no

Key result

Type:

density

Density:

ca. 1.62 g/cm³

Temp.:

20 °C

Conclusions:

4.6. Density
Computational tools: T.E.S.T./ Hierarchical Clustering
Density of the target 2,4,8,10-Tetraoxa-3,9-dithiaspiro[5.5]undecane, 3,3,9,9-tetraoxide was estimated by
using a battery of tools, including ACD/Percepta Physchem profiler and four T.E.S.T. models (Hierarchical
Clustering, Group Contribution, FDA and Nearest neighbor). Following a detailed assessment of the
generated QSAR predictions, the most reliable prediction was provided by T.E.S.T./ Hierarchical clustering
model.Business confidential – (Q)SAR predictions for regulatory purposes
24/11/2020 35
The Hierarchical Clustering model implemented in T.E.S.T. estimates density using the weighted average of
the predictions from several different models, which are trained on clusters of structurally similar training set
compounds. The prediction results are illustrated in Table 4.9 and following discussed.
Table 4.9 Density prediction.
Tool Density (g/cm3) Reliability assessment
Predicted value Prediction interval
TEST/Hierarchical clustering 1.62 1.59 – 1.66 MODERATE RELIABLE
Prediction Density equal to 1.62 g/cm3 (considering the uncertainty interval, density ranging from 1.59 to 0.66 g/cm3).
Prediction based on four cluster models, providing the following predictions: 1) 1.71 ± 0.07;
2) 1.76 ± 0.07; 3) 1.58 ± 0.04; 4) 1.60 ± 0.06.
Applicability
Domain
(AD)
The applicability domain of the model is defined using several different constraints. 1) The
model ellipsoid constraint checks if the target chemical is within the multidimensional
ellipsoid defined by the ranges of descriptor values for the chemicals in the cluster (for the
descriptors appearing the cluster model). The model ellipsoid constraint is satisfied if the
leverage of the target compound is less than the maximum leverage value for all the
compounds used in the model. 2) The Rmax constraint checks if the distance from the target
chemical to the centroid of the cluster is less than the maximum distance for any chemical in
the cluster to the cluster centroid. The distance is defined in terms of the entire pool of
descriptors (not only the descriptors appearing in the model). 3) The fragment constraint is
that the compounds in the cluster have to have at least one example of each of the fragments
contained in the target chemical. If a chemical contains a fragment that is not present in the
training set, no prediction is provided.
The target compound 2,4,8,10-Tetraoxa-3,9-dithiaspiro[5.5]undecane, 3,3,9,9-tetraoxide is
included in the model applicability domain of the four claster models since none of the three
constrains, i.e. model ellipsoid, Rmax and fragment constraints, were violated.
• descriptor domain: included (no constrains violated).
• structural fragment domain: included (no constrains violated).

The analysis of uncertainties and reliability of the prediction provided by T.E.S.T. was
performed considering the following aspects: i) target prediction interval; ii) applicability
domain considerations; iii) degree of similarity of training/test set analogues; iv) estimation
accuracy of training/test set analogues, based on the MAE (Mean absolute error in g/cm3)
provided by the program for the chemicals with a similarity coefficient ≥ 0.5.
For the target 2,4,8,10-Tetraoxa-3,9-dithiaspiro[5.5]undecane, 3,3,9,9-tetraoxide, the QSAR
prediction was assessed as moderately reliable (limited uncertainty) based on the following
considerations:
• limited prediction interval was calculated for the target, leading to a density
prediction falling in the following range: 1.59-1.66.
• applicability domain: the target is included in the AD of the model.
• structural similarity of training/test set analogues: limited to moderate
• prediction accuracy of training/test set analogues: good.
Adequacy The density QSAR prediction was assessed as adequate for regulatory purposes (Klimisch 2
– results derived from a valid QSAR model and falling into its applicability domain, with
adequate and reliable documentation/justification)

Executive summary:

Density equal to1.62 g/cm3

For the target 2,4,8,10-Tetraoxa-3,9-dithiaspiro[5.5]undecane, 3,3,9,9-tetraoxide, the QSARprediction was assessed as moderately reliable (limited uncertainty) based on the following considerations:

• limited prediction interval was calculated for the target, leading to a density prediction falling in the following range: 1.59-1.66.

Adequacy The density QSAR prediction was assessed as adequate for regulatory purposes (Klimisch 2– results derived from a valid QSAR model and falling into its applicability domain, with adequate and reliable documentation/justification)

Description of key information

 

Density equal to1.62 g/cm3(considering the uncertainty interval, density ranging from 1.59 to 0.66 g/cm3).

 

Key value for chemical safety assessment

Relative density at 20C:

1.62

Additional information

QSAR