Sodium 3-nitrobenzenesulphonate

Adsorption / desorption

Link to relevant study record(s)

Reference

Endpoint:

adsorption / desorption, other

Type of information:

calculation (if not (Q)SAR)

Adequacy of study:

weight of evidence

Reliability:

4 (not assignable)

Rationale for reliability incl. deficiencies:

secondary literature

Justification for type of information:

Data is from Comptox Chemistry Dashboard database

Qualifier:

according to guideline

Guideline:

other: Refer below principle

Principles of method if other than guideline:

Prediction done using OPERA (OPEn (quantitative) structure-activity Relationship Application) V1.02 model in which calculation based on PaDEL descriptors (calculate molecular descriptors and fingerprints of chemical)

GLP compliance:

not specified

Type of method:

other: PaDEL descriptors

Media:

soil

Radiolabelling:

not specified

Test temperature:

No data

Analytical monitoring:

not specified

Key result

Type:

Koc

Value:

25.7 L/kg

Remarks on result:

other: log Koc = 1.409, Result based on the OECD principle 1-5

Transformation products:

not specified

Prediction based on following 5 OECD principles:

OECD Principle 1 (Defining the endpoint):

The original data collected from the PHYSPROP database (788 chemicals) have undergone a series of processes to curate the chemical structures and remove duplicates, obvious outliers and erroneous entries. This procedure also included a consistency check to ensure only good quality data is used for the development of the QSAR model (750 chemicals).

 

Then, QSAR-ready structures were generated by standardizing all chemical structures and removing duplicates, inorganic and metallo-organic chemicals (735 chemicals). The descriptions of KNIME workflows that were developed for the purpose of the cleaning and standardization of the data are available in the papers [ref 1 and ref 4 Section 2.7].

 

The curated outlier-free experimental data (729 chemicals) was divided into training and validation sets before the machine learning and modeling steps.

 

OECD Principle 2(Defining the algorithm):

Type of model:

QSAR model using PaDEL descriptors

 

Explicit algorithm:

Distance weighted k-nearest neighbors (kNN) This is a refinement of the classical k-NN classification algorithm where the contribution of each of the k neighbors is weighted according to their distance to the query point, giving greater weight to closer neighbors.The used distance is the Euclidean distance. kNN is an unambiguous algorithm that fulfills the transparency requirements of OECD principle 2 with an optimal compromise between 4.Defining the algorithm - OECD Principle 2 model complexity and performance.

 

OECD Principle 3(Defining the applicability domain):

Method used to assess the applicability domain:The applicability domain of the model is assessed in two independent levels using two different distance-based methods. First, a global applicability domain is determined by means of the leverage approach that checks whether the query structure falls within the multidimensional chemical space of the whole training set. The leverage of a query chemical is proportional to its Mahalanob is distance measure from the centroid of the training set. The leverages of a given dataset are obtained from the diagonal values of the hat matrix. This approach is associated with a threshold leverage that corresponds 5.Defining the applicability domain - OECD Principle 3 to 3*p/n where p is the number of model variables while n is the number of training compounds. A query chemical with leverage higher than the threshold is considered outside the AD and can be associated with unreliable prediction. The leverage approach has specific limitations, in particular with respects to gaps within the descriptor space of the model or at the boundaries of the training set. To obviate such limitations, a second tier of applicability domain assessement was added. This comprised a local approach which only investigated the vicinity of the query chemical. This local approach provides a continuous index ranging from 0 to 1 which is different from the first approach which only provides Boolean answers (yes/no). This local AD-index is relative to the similarity of the query chemical to its 5 nearest neighbors in the p dimensional space of the model. The higher this index, the more the prediction is likely to be reliable.

 

OECD Principle 4 (Internal validation):

Availability of the training set: Yes

Statistics for goodness-of-fit: Performance in training: R2=0.81 RMSE=0.54

Robustness - Statistics obtained by leave-many-out cross-validation: Performance in 5-fold cross-validation: Q2=0.81 RMSE=0.55

 

OECD Principle 4(External validation):

Availability of the external validation set: Yes

.Predictivity - Statistics obtained by external validation: Performance in test: R2=0.71 RMSE=0.61

Experimental design of test set:

The structures are randomly selected to represent 25% of the available data keeping a similar normal distrubution of LogKoc vlaues in both training and test sets using the Venetian blinds method.

 

OECD Principle 5 (Providing a mechanistic interpretation):

Mechanistic basis of the model: The model descriptors were selected statistically but they can also be mechanistically interpreted. KOC is the ratio between the concentration of a chemical adsorbed by the soil normalized to soil organic carbon and the concentration dissolved in the soil water. Thussoil sorption is closely related to water solubility and logP. Therefore, the chemical features which determine the soil sorption are similar to those related to water solubility and logP. In particular, size related descriptors since larger compounds tend to have higher soil sorption because they do have lower water solubility. Also elctronic profile descriptors related to charges and to charge distribution are of high importance: the presence of active functional group next to carbon leads to better water solubility, likewise higher polarity leads to better water solubility.

Validity criteria fulfilled:

not specified

Conclusions:

From CompTox Chemistry Dashboard using OPERA (OPEn (quantitative) structure-activity Relationship Application) V1.02 model in which calculation based on PaDEL descriptors (calculate molecular descriptors and fingerprints of chemical) the adsorption coefficient i.e KOC for test substance was estimated to be 25.7 L/kg .

Executive summary:

From CompTox Chemistry Dashboard using OPERA (OPEn (quantitative) structure-activity Relationship Application)  V1.02 model in which calculation based on PaDEL descriptors (calculate molecular descriptors and fingerprints of chemical), the adsorption coefficient i.e KOC for test substance was estimated to be 25.7 L/kg (log Koc = 1.409).The predicted KOC result based on the 5 OECD principles. This Koc value indicates that the test substance has a negligible sorption to soil and sediment and therefore have rapid migration potential to ground water.

Description of key information

From CompTox Chemistry Dashboard using OPERA (OPEn (quantitative) structure-activity Relationship Application)  V1.02 model in which calculation based on PaDEL descriptors (calculate molecular descriptors and fingerprints of chemical), the adsorption coefficient i.e KOC for test substance was estimated to be 25.7 L/kg (log Koc = 1.409).The predicted KOC result based on the 5 OECD principles. This Koc value indicates that the test substance has a negligible sorption to soil and sediment and therefore have rapid migration potential to ground water.

.

Key value for chemical safety assessment

Koc at 20 °C:

25.7

Additional information

Predicted data for the target compound and supporting experimental study for its read across substances were reviewed for the adsorption end point which is summarized as below:

From CompTox Chemistry Dashboard using OPERA (OPEn (quantitative) structure-activity Relationship Application) V1.02 model in which calculation based on PaDEL descriptors (calculate molecular descriptors and fingerprints of chemical), the adsorption coefficient i.e KOC for test was estimated to be 25.7 L/kg (log Koc = 1.409).The predicted KOC result was based on the 5 OECD principles.

In a supporting study from authoritative database (HSDB, 2017) for the test chemical, adsorption experiment was conducted for estimating the adsorption coefficient (Koc) value of test chemical. The adsorption coefficient (Koc) value was calculated using an experimental water solubility of 80,000 mg/l and a regression derived equation. The adsorption coefficient (Koc) value of test chemical was estimated to be 9 (Log Koc = 0.954). This Koc value indicates that the test chemical has a negligible sorption to soil and sediment and therefore has rapid migration potential to ground water.

In next supporting study the Koc of test chemical was estimated to be approximately 54 , the log Koc was calculated to be 1.732 . Hence, based on the log Koc value the test chemical is considered to have Low sorption to soil and sediment, moderate migration to ground water.

On the basis of above overall results for target chemical (from modelling database CompTox Chemistry Dashboard, 2017and authoritative database HSDB, 2017), it can be concluded that the Koc value of test chemical was estimated to be 9 to 54 indicating that the test chemical has a negligible to low sorption to soil and sediment and therefore have rapid to moderate migration potential to ground water.

[LogKoc: 1.409]