Lära Introduction to Molecular Representations | Molecular Representations and Parsing

Förkunskaper

Introduction to Python

Digital molecular representations are essential in chemoinformatics because they allow you to store, search, and analyze chemical structures efficiently using computers. Instead of relying on graphical depictions or physical models, you can use text-based formats to represent molecules in a way that is easy for both humans and machines to process. Two of the most widely used representations are SMILES and InChI. These formats enable you to encode the structure of molecules as strings of characters, making it possible to perform tasks like molecular comparison, database searching, and computational modeling at scale.


              1234567891011121314
            
# Example SMILES and InChI strings for simple molecules

# Ethanol
smiles_ethanol = "CCO"
inchi_ethanol = "InChI=1S/C2H6O/c1-2-3/h3H,2H2,1H3"

# Benzene
smiles_benzene = "c1ccccc1"
inchi_benzene = "InChI=1S/C6H6/c1-2-4-6-5-3-1/h1-6H"

print("Ethanol SMILES:", smiles_ethanol)
print("Ethanol InChI:", inchi_ethanol)
print("Benzene SMILES:", smiles_benzene)
print("Benzene InChI:", inchi_benzene)

The structure of a SMILES string encodes the atoms and bonds in a molecule using a specific set of rules. Atoms are represented by their atomic symbols, while single, double, and triple bonds are usually implied or shown with special characters. Branches and ring closures are also indicated with specific syntax. For example, in the SMILES string "CCO" for ethanol, each C is a carbon atom and O is an oxygen atom, with single bonds implied between them.

InChI (International Chemical Identifier) is a more systematic and layered representation. It starts with "InChI=1S/" followed by information about the molecule's formula, connectivity, hydrogen atoms, and sometimes stereochemistry. InChI strings are longer and less human-readable, but they are designed to be unique and unambiguous for each molecule, which makes them useful for database indexing and interoperability.


              123456789101112131415161718
            
# Manually interpreting a SMILES string

smiles = "CCO"  # Ethanol

# Breakdown:
# The first "C" represents a carbon atom (the methyl group).
# The second "C" is another carbon atom (the methylene group), bonded to the first.
# The "O" is an oxygen atom (the hydroxyl group), bonded to the second carbon.
# All bonds are single unless otherwise specified.

atoms = ["C", "C", "O"]
bonds = [
    ("C1", "C2", "single"),
    ("C2", "O", "single")
]

print("Atoms in ethanol:", atoms)
print("Bonds in ethanol:", bonds)

SMILES offers the advantage of being compact and relatively easy to read for simple molecules, which makes it popular for data entry, visualization, and substructure searching. However, it can be ambiguous for complex molecules without strict canonicalization, and different software may generate different SMILES for the same molecule.

InChI, in contrast, is designed to provide a unique identifier for each molecule, making it ideal for database indexing and ensuring that different sources refer to the same compound. Its layered format captures more detailed structural information, but it is less human-readable and more difficult to use for quick visual inspection or manual editing. You will often use SMILES for tasks like drawing molecules or searching for substructures, and InChI when you need precise, standardized identifiers for data sharing or publication.

Definition

Definition:

SMILES stands for "Simplified Molecular Input Line Entry System." It is a line notation for describing chemical structures using short ASCII strings. SMILES is commonly used for representing molecules in chemical databases, drawing tools, and cheminformatics software.
InChI stands for "International Chemical Identifier." It is a non-proprietary, textual identifier developed by IUPAC to provide a standard way to encode molecular information and facilitate the search for such information in databases and on the web. InChI is often used for database indexing, interoperability, and ensuring unique identification of molecules.

Var allt tydligt?

Tack för dina kommentarer!

Avsnitt 1. Kapitel 1

Fråga AI

Fråga vad du vill eller prova någon av de föreslagna frågorna för att starta vårt samtal

Svep för att visa menyn