liveBook · Manning

SMILES

As explained in Build AI Drug Discovery Pipelines

SMILES is a line notation that encodes molecular atoms, bonds, and connectivity as a text string using defined symbols and grammar rules.

SMILES (Simplified Molecular Input Line Entry System) encodes molecular atoms, bonds, and connectivity as a compact text string. This representation gives machine-learning workflows a practical input for molecular property prediction, virtual screening, and related drug-discovery tasks. section 1.2 section 1.5

Read a SMILES string using its vocabulary and grammar:

  • Write atoms with element symbols such as C, N, O, Cl, and Br; bracket atoms when special valences, attached hydrogens, or formal charges require explicit notation. section 1.2.1
  • Represent single, double, triple, and aromatic bonds with -, =, #, and :; single and aromatic bonds may be implicit. section 1.2.1
  • Use parentheses for branches, numerical markers for ring closures, and lowercase atom symbols for aromatic structures. section 11.2.2
  • Treat equivalent strings carefully: one compound can have multiple valid SMILES, so a string is not always a unique textual identity. section 1.2.1

Aspirin as a molecular graph

The SMILES string "CC(=O)Oc1ccccc1C(=O)O" describes aspirin. Reading left to right reveals a methyl group, carbonyl branch, oxygen-linked aromatic ring, ring closure, and carboxyl group. RDKit can parse this string, validate the structure, extract atoms and bonds, and produce a graph representation for graph neural networks. section 11.2.2

Table 1.1 Examples of SMILES. A complete reference on SMILES can be found in the Daylight manual.
SMILES
Name
CC
Ethane
O=C=O
Carbon Dioxide
C#N
Hydrogen Cyanide
CCN(CC)CC
Triethylamine
CC(=O)O
Acetic Acid
C1CCCCC1
Cyclohexane
c1ccccc1
Benzene

Use the example table to compare common SMILES strings with named compounds, including ethane, carbon dioxide, acetic acid, cyclohexane, and benzene. section 1.2.1

How Build AI Drug Discovery Pipelines covers it

How deep it goes
Explains symbols, bonds, branches, rings, stereochemistry, equivalent strings, and RDKit graph conversion, but defers complete grammar details.
Examples
Demonstrates aspirin, methane, ethane, carbon dioxide, hydrogen cyanide, acetic acid, cyclohexane, benzene, and triethylamine.
What it assumes
Assumes basic molecule, atom, and covalent-bond vocabulary while introducing SMILES notation from first principles.
What you can do after
Build molecular graph inputs from SMILES using RDKit, including atom features, bonds, directed edges, and self-loops.
How it connects
Connects molecular graphs, machine learning, graph neural networks, drug discovery, and molecular property prediction.

Also explained in

Compare how 2 books explain SMILES →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Build AI Drug Discovery Pipelines. The section links point to the parts of the book it draws on.