[4-[[4-anilino-1-naphthyl][4-(dimethylamino)phenyl]methylene]cyclohexa-2,5-dien-1-ylidene]dimethylammonium chloride

Skin sensitisation

Administrative data

Endpoint:

skin sensitisation, other

Remarks:

QSAR prediction

Type of information:

(Q)SAR

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

accepted calculation method

Justification for type of information:

SOFTWARE: QSAR DYES R&C
Report produced by version QSAR dyes R&C 2.1
Further details are available in the attached document.

Data source

Reference

Reference Type:

other: QSAR report

Title:

Unnamed

Year:

2017

Materials and methods

Principles of method if other than guideline:

QSAR prediction. Details on the QSAR model are available in the attachment, i.e. QSAR model reporting (QMRF).

GLP compliance:

no

Type of study:

other: QSAR prediction

Test material

Test material information

Constituent 1

Reference substance name:

Basic Blue 026 chloride

IUPAC Name:

Basic Blue 026 chloride

Results and discussion

In vitro / in chemico

Results

Parameter:

other: QSAR prediction

Remarks on result:

positive indication of skin sensitisation

Applicant's summary and conclusion

Interpretation of results:

other: skin sensitising according to the CLP Regulation (EC 1272/2008)

Conclusions:

Skin sensitising potential, based on a prediction by a suitable QSAR model.

Executive summary:

Method

QSAR prediction. The skin sensitisation prediction of the target molecule is obtained by the k-Nearest Neighbours (kNN) strategy [Kowalski 1972], which is a well-known non-parametric classification method based on the concept of similarity. The predicted value for a molecule is usually computed on the basis of the frequency (or weighted frequency) of the experimental classes of its k nearest neighbours. In the skin sensitisation QSAR model, the sensitisation effects (sensitising or not sensitising) of the nearest molecules are used to predict the target sensitisation and to calculate a reliability score of the prediction. The kNN approach has been implemented with a variable k parameter and the prediction has been calculated as the similarity weighted average of the responses of the k nearest molecules to allow the most similar molecules to mainly contribute to the prediction. The reliability score is a measure of the prediction reliability. It ranges from 0 to 1: 1 indicates that all the nearest molecules belong to the same experimental class and are very similar to the target. Note that along with the reliability score also the number of similar structures used for prediction should be accounted for evaluating the prediction confidence.

Results

Prediction was in applicability domain. Target molecule resulted as skin sensitising.