Tutorial: Reaction SMARTS tutorial

Each MITE entry describes the reaction and substrate specificity of the tailoring enzyme.

To represent this information in a format that is both human- and machine-readable, reaction SMARTS and SMILES are used. Below, you can find more information about this format, including a step-by-step tutorial.

What are SMILES, SMARTS, and reaction SMARTS?

SMILES and SMARTS are line notations used to describe chemical structures. SMILES string represents a molecule by specifying its atoms and their connectivity, while a SMARTS string defines a search pattern of atoms and bonds.

For example, the SMILES string C=C represents ethylene, while the SMARTS string [C]=[C] matches both ethylene and propylene (C=CC).

Reaction SMARTS extend this concept by modeling chemical reactions, encapsulating the pattern matching of substrates and their conversion to products (substrate(s)>>product(s)). This system allows for the modeling of reactions and their application to SMILES strings.

In MITE, reaction SMARTS describe the substrate specificity and reactions of tailoring enzymes, while SMILES are used to represent the actual substrates and products of characterized reactions.

How to create reaction SMARTS and SMILES?

Writing SMILES and reaction SMARTS by hand can be complex and error-prone. Instead, we recommend drawing the chemical structures first and then exporting them as string notations.

Although there are many chemistry drawing tools available, we recommend the free, open-source tool Ketcher, which offers an online demo version and will be used in this tutorial.

Step-by-step guide

Note

Instructions can be also found in this video tutorial.

In this example, we will characterize the reaction of the enzyme PltM, a phenol halogenase. We will draw the reaction in its easiest form and export the SMILES and reaction SMARTS. In the publication, PltM is described to halogenate the phloroglucinol substrate, resulting in a mono- or di-halogenated product, with either chloride, bromide, or iodide substituents. Here, we will model the mono-chlorination reaction.

1. In Ketcher, draw phloroglucinol and convert it into its aromatic form by pressing ALT+A. As a rule of thumb, all molecules must be turned into their aromatic (not kekulized) form to be MITE-compatible.

2. Select phloroglucinol by drawing a rectangle (or pressing CTRL+A) and export it as a Daylight SMILES string using the Save Structure menu (pressing CTRL+S). c1c(O)cc(O)cc1O is the SMILES string of the substrate.

3. In another Ketcher window, draw the chloro-phloroglucinol product. You can draw the structure from scratch or copy-paste phloroglucinol from the first Ketcher window (CTRL+C/CTRL+V) and add a chloride substituent. Do not forget to aromatize as before with ALT+A. As before, select the molecule by drawing a rectangle (or pressing CTRL+A) and export it as a Daylight SMILES string using the Save Structure menu (pressing CTRL+S). c1c(O)cc(O)c(Cl)c1O is the SMILES string of the product.

4. Select the chloro-phloroglucinol and copy it (CTRL+C). Go back to the first window containing phloroglucinol, add a reaction arrow, and paste the chloro-phloroglucinol on the right-hand side of the arrow. This is your reaction.

5. Next, select the reaction (CTRL+A), and select the Reaction Auto Mapping Tool, using the “Discard” mode. This will add paired atom indices to all atoms that are both on the substrate and the product side. Since the chloride atom is only present on the product side, it does not receive an index.

6. Select the reaction again (CTRL+A) and export it as a Daylight SMARTS (CTRL+S). [#6:1]1:[#6:5](-[#8:9]):[#6:6]:[#6:4](-[#8:8]):[#6:2]:[#6:3]:1-[#8:7]>>[#6:1]1:[#6:5](-[#8:9]):[#6:6]:[#6:4](-[#8:8]):[#6:2](-Cl):[#6:3]:1-[#8:7] is the reaction SMARTS string. With substrate and product SMILES and reaction SMARTS prepared, you can test-run this reaction.

7. Fill in the reaction SMARTS and the example substrate-product pair into the form field, and press on Test reaction. If the reaction runs without errors, it can be entered in the MITE submission form during entry creation.

Examples

Example for minimal amino acid sequence

see MITE0000406.

Author(s) of this page: Marcel Fröhlich, Susanna Kummer, Mitja M. Zdouc