Using the VEGA platform, you can access a series of QSAR (quantitative structure-activity relationship) models for regulatory purposes, or develop your own model for research purposes.
QSAR models can be used to predict the property of a chemical compound, using information obtained from its structure.
More information about QSAR models is available here
The QSAR models available for regulatory purposes have been optimised in accordance with the REACH requirements.
The models have been taken from CAESAR or T.E.S.T., or have been developed later by the contributors to VEGA.
Compared with many existing QSAR models, we have put greater emphasis on ensuring that the models generate transparent, understandable, reproducible and verifiable results. To achieve this, a series of tools has been optimised, which can relate the results obtained for the target chemical to the results obtained for similar (structurally related) compounds.
The tools therefore also provide a reproducible procedure for read-across. Read-across is a tool to extract the required values for the target compound on the basis of the known values for similar (structurally related) substances. Within VEGA this has been achieved through the implementation of algorithms that are independent of the QSAR models, and which allow the screening of similar compounds in a read-across strategy. These algorithms are used both to identify similar compounds, and to analyse the role of descriptors and fragments both for the chemical of interest and for those similar compounds.
It is therefore also possible to use VEGA solely for read-across. For this more limited use, you proceed as described for the prediction procedure, but then disregard the predicted value, and instead simply analyse the similar compounds on the chemometric basis (step 4), and the similar compounds identified on the basis of the mechanisms and descriptors (step 5, Descriptors and fragments). In this way you will analyse and review the similarities between the target chemical and the related chemicals.
