Usage

The Skin sensitization ITS predictor can be used through the web interface, using APIs or through integration into your workflow of choice (e.g., KNIME or Pipeline Pilot).

The web interface

You can try the skin sensitization integrated testing strategy for free on this demo page.
The predictor interface is intuitive and allows the risk assessor to combine information from in silico, in chemico and in vitro assays to reach a comprehensive judgement on the skin sensitization potential of a chemical.

Step 1

Enter the molecule of interest either by sketching the molecule or entering a SMILES representation. You can also select one of the predefined examples from the drop-down menu as seen in figure 1.

Step 1: Submit molecule
Figure 1. Submit molecule as SMILES, by sketching or select a predefined example.

Step 2

The in silico descriptors will be calculated based on our underlying QSAR models. If descriptor values were not cached before, they will be calculated on the spot as shown in figure 2. Please allow for few seconds for all calculations to complete before proceeding to the next step.

Step 2: In silico descriptors
Figure 2. In the second step, in silico descriptors for bioavailability and reactivity are being calculated.

After calculations are complete, you might want to check for the computed estimates. Such values can be edited if you believe that you have better estimates (for example, experimental values for water solubility).

Step 3

Enter the experimental values from the validated alternative assays, DPRA, KeratinoSens and h-CLAT as shown in figure 3. Clicking on the top right corner of each assay provides more information about the assay and the expected input values.

Step 3: Assay values
Figure 3. The third step where users enter experimental values.

Note: Based on the compound’s solubility and its fraction ionized at pH = 7, a judgement is made to whether a chemical is considered within the applicability domain of a given assay. If it was not, the assay values are ignored and the corresponding input textbox turns orange (as shown in figure 4) as an indication that the chemical is out of applicability domain.

Alert if chemical is out of assay’s applicability domain
Figure 4. Users receive alerts when chemical is out of assay’s applicability domain.

Step 4

Finally, click on “Make prediction” as shown in figure 5. The network returns the estimation of compound’s skin sensitization (pEC3 class) as well as an estimate of the confidence in prediction based on the Bayes factor for the predicted class (figure 6).

Click to make a prediction
Figure 5. Click on “Make prediction” button
Prediction result
Figure 6. The result

Input parameters

The DC ITS SkinSens integrates information from multiple sources for a holistic assessment of exposure hazard. Below is the list of the values used for making the predictions:

Experimental values

  • DPRACYS, DPRALYS - Experimental value from key event 1: % of the cysteine-(Cys) peptide remaining in the DPRA assay (Key Event 1: peptide reactivity [OECD 442 C: Direct peptide reactivity test (DPRA) (OECD 2015c)])
  • KEC1.5,KEC3, IC50 - Experimental values from Key Event 2: KeratinoSens™ Concentration yielding 1.5-fold (KEC1.5); threefold (KEC3) induction of Nrf2-dependent luciferase activity in the KeratinoSens™ assay; 50 % reduction in cell viability in the KeratinoSens™ assay
  • EC150, EC200, CV75 - Experimental values from Key Event 3: Concentrations yielding 150 % induction of the cell surface activation marker CD86 in the h-CLAT; 200 % induction of the cell surface activation marker CD54 in the h-CLAT; 25 % reduction in cell viability in the h-CLAT

In silico descriptors

  • TIMES - The highest skin sensitization class [1-3] for the compound and its potential metabolites as predicted by TIMES in silico model.
  • Water solubility @ pH7 - Predicted using ChemAxon calculator.
  • Logkow - Octanol/Water partition coefficient, calculated using AlogPS 3.01 from OCHEM web services.
  • Protein binding - Plasma protein binding percentage as predicted using QSAR model published on https://ochem.eu/model/12651495.
  • logD @ pH7 - Partition coefficient at pH = 7 as predicted by ChemAxon calculator.
  • Michael acceptor - If a compound is a Michael acceptor, a post-prediction correction is performed to accomodate for the anti-inflammatory effect of such chemicals.

API

DC ITS SkinSens offers an API (Application programming interface) intended to be used as an interface by software components to communicate with each other. The Skin sensitization ITS API can be accessed as passing the in vitro and in silico parameters for prediction as query parameters. Below is an example for the prediction of sensitization of dimethyl formamide:


https://its-r-predict.cloud.douglasconnect.com/predict?CV75=68404.7867&DPRACYS=99.5&DPRALYS=100&EC150=10000&EC200=10000&IC50=2028&KEC1.5=2026&KEC3=1512&LOG.D.PH7=-0.829&LOGKOW=-1.01&MW=73.094&PROTEIN.BINDING=16.06&TIMES=1&WSPH7.M.L=13.681&smiles=CN(C)C=O
        

The results are returned as a json response.

KNIME

Skin Sensitization ITS will also be available as a node in the KNIME workflow management system. The node is currently planned for release by Winter 2018. For early access to our KNIME node, please contact us.

Attribution

JSME Editor
Copyright (c) 2017, Novartis Institutes for BioMedical Research Inc and Bruno Bienfait
All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
  • Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
  • Neither the name of the Novartis Institutes for BioMedical Research nor the names of the authors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL COPYRIGHT HOLDERS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.