Molecular Fingerprint

This is a TypeScript package for generating Morgan fingerprints from chemical SMILES notation.

Overview

Morgan fingerprints are circular fingerprints that represent the structural features of a molecule. This class provides functionality to convert a given SMILES notation into Morgan fingerprints, which can be used for tasks like molecular similarity comparison, substructure searching, and machine learning in cheminformatics.

Installation

Install molecular-fingerprint with npm

  npm install molecular-fingerprint

Usage

To use this class, simply import it into your TypeScript project and call the convertToMorganFingerprint method, passing the SMILES notation of the molecule you want to generate fingerprints for.

import { MolecularFingerprint } from 'molecular-fingerprint'

const smiles = 'C1=CC=CC=C1'
const fingerprint = MolecularFingerprint.convertToMorganFingerprint(smiles)
console.log(fingerprint)

Optionally, you can specify the radius parameter for the Morgan fingerprint. The default value is 2.

const radius = 3
const fingerprint = MolecularFingerprint.convertToMorganFingerprint(smiles, radius)
console.log(fingerprint)

Parameters

SMILES: The Simplified Molecular Input Line Entry System (SMILES) notation of the molecule.

Radius: The radius parameter controls the number of atoms considered in the substructure generation. Default value is 2.