Create a SAR Slides from a Scaffold

Intro

In this tutorial we'll guide you through the steps of creating a SAR Slide based on a scaffold. We'll use a demo dataset (that you can download here - HSD17B13 DATASET{target=_blank}). In case you did not already, you can follow the tutorial showing how to import the dataset in the application..

1. Select your starting dataset

From the home page of the application, click on “SAR Slide” and select “New SAR Slide” in the next page.

Select the HSD17B13 dataset. You should end up with the following form:

2. Draw your scaffold

• Click on the "Scaffold" tab. An empty sketcher will be shown.
• Type in "34" in the Load compound in sketcher select box on the top and select it. Note that you could use suggested scaffolds instead (see Select Starting points for SAR Reports{target=_blank}), but the AIM of this tutorial is to explain how to draw / edit scaffolds using our sketcher.

Substition points

Let's now draw substitution points we'll be interested in. To do so, you first have to remove extra atoms that are not part of the scaffold you want to submit.

There are various ways of removing atoms:

• When hovering on any atom / bond, click on the Delete key in your keyboard. Note that you can also keep the delete key pressed and hover on atoms / bonds you want to remove.
• Click on the Eraser button (second button on the top left bar of the sketcher), and then click or select atoms to be removed.

Remove atoms to obtain the following structure:

Once done, let's define positions representing substitution points. Again, two options:

• Hover over the targeted atom, and press the R key on your keyboard and type enter on your keyboard (to validate the label - before typing enter, you can eventually assign a different number to your r-group).
• Use the R-Group button "R" available on the left vertical sidebar: click on it, and then click on every atom you want to define as an attachment point. After clicking, the R-Group label is automatically assigned - you can edit it eventually - and you'll need to press enter on your keyboard to validate the label.

In both cases, remember to press enter to validate the R-Group labeling. You should end up with the following scaffold:
**

Make the scaffold fuzzy

Various series exist in this dataset, with the central thiazole ring being replaced by analogs. Let's make our scaffold fuzzy to retrieve these as well by assigning "any atom" query feature to our scaffold:

• Hover over on the sulfur atom with your mouse, and press the star key on your keyboard.
• Repeat the operation for the nitrogen atom and the carbon with no R-Group attached.

This should be the final state of our scaffold:

• Submit the calculation by clicking on the Generate SAR Slide button.

After a few seconds of processing, you should land in a new SAR Slide page showing the scaffold you drawn along with the R-Groups detected at each positions.

3. Interpreting and navigating scaffold-based SAR Slides

The resulting SAR Slides shows, at the center, the scaffold you drawn. Attached to each R-Group, a colored bubble lists all fragments that exist at that specific position of the scaffold. Since we made our scaffold fuzzy, the scaffold itself also shows up as a varying part with a dedicated gallery:

At a given R-Group position, a fragment may be present in more than one molecule (as opposed to compound-centric SAR Slides based on MMPs that shows individual transformation between pairs of compounds). We show by default the identifier and property values of the compound that has the best "Main Endpoint" activity. The fragments are also ordered based on the compound with the best associated activity.

Per-fragment activity distribution

In R1 position, we can see that the first fragment has 11 compounds associated with it. This difluorophenol is also associated with the most active compound among all other molecules matching our query. Let's have a closer look:

• On the R1 gallery, hover on the Enzymatic pIC50 activity value bellow the first difluorophenol image, and click on the value. An histogram should appear attached to the value. Repeat the operation for the Cellular pIC50.

These histograms show the distribution of the selected property values of molecules having this difluorophenol in R1 (red), compared to the distribution of the activities for the entire dataset. We can easily see here that most compounds having that particular difluorophenol are associated with high pIC50.

Filtering based on R-Group(s)

Let's focus the report on this particular subset of molecules:

• Hover on the fragment image. You should see a filter icon appear on the top right corner of the image.

• Click on the filter icon. All other galleries will be updated accordingly, showing only fragments associated with the filter we just applied:

Reviewing existing / missing combinations

• Focus on the R1 gallery, and click on the "11 compounds" label shown bellow the fragment. It will open a table view showing molecules corresponding to this fragment:

• Click on the Matrix tab on the top of the modal window. You'll end up in an alternative view showing a 2D SAR Matrix.

This view shows, by default, the existing combinations between the Scaffold part and R2 part. A coloring scheme is automatically applied, and the fragments on each axis are order based on the "best" compound they are associated with. You therefore should see the best combinations by default. Cells that are not colored correspond to combinations that have not been made so far.