MolecularGraph.jl

MolecularGraph.jl is a graph-based molecule modeling and chemoinformatics analysis toolkit fully implemented in Julia.

Documentation and API Reference

To run codes in your environment, see Edit or run this notebook instruction shown in the top-right of the tutorial pages below.

Getting started
Molecular graph basics
- Scope of MolecularGraph.jl
- Considerations in molecular graph implementation
- Basic operations provided by Graphs.jl interface
- MolGraph type and atom/bond properties
Properties and descriptors
- Built-in molecule properties and descriptors
  - Lipinski's Rule of five (RO5)
  - Molecular formula
  - Atom and bond properties
  - Graph topology (ring and fused ring)
- Auto-update mechanism of properties
Preprocessing
- Remove hydrogen vertices
- Extract molecules of interest
- Standardize charges
- Dealing with resonance structure
- Customize property updater
Mass and isotopes
- Molecular weight and exact mass
- Uncertainty
- Isotopic composition
- Simulate mass spectrum
Substructure and query
- Substructure match
- InChI and InChIKey
- SMARTS query
- Structural alerts (e.g. PAINS)
- Functional group analysis
- Query containment
Maximum common substructure (MCS)
- Maximum common induced substructure (MCIS)
- Maximum common edge-induced substructure (MCES)
- Connected or disconnected MCS
- Working with larger molecules
- Topological constraint (tdMCS)
Drawing molecule
- 2D structure images
  - Display and export in SVG format
  - Display and export in PNG format
  - Layout for Pluto notebook
  - Regenerate coordinates
- 3D molecule rendering using Makie.jl
Stereochemistry
- Stereochemistry as a molecular graph property
- Stereospecific implicit hydrogens

Chemical structure file I/O
- 2D structure image drawing and export to SVG
- 3D structure drawing (Makie.jl)
- SDFile reader/writer (.sdf, .mol)
- SMILES/SMARTS parser (only reader)
- Coordinates generation (coordgenlibs)
Properties and descriptors
- H-bond donor/acceptor
- rotatable bonds
- Aromaticity
- Wildman-Crippen logP
Substructure and query
- InChI (InChI)
- Serialization (molecule object <-> JSON)
- Subgraph isomorphism detection with VF2 algorithm
  - SMARTS query match
  - Monomorphism, node-induced and edge-induced match
  - Constraints (mandatory/forbidden vertex mapping)
- Functional group query set
- Structural alerts detection with ChEMBL dataset
Atomic mass
- standard atomic/molecular weight
- relative atomic/molecular mass
- isotopic composition
Molecular graph topology
- Ring, scaffold, connected components
- Minimum cycle basis (de Pina algorithm)
  - Smallest set of smallest rings (SSSR)
- Planarity (left-right planarity test)
- Maximum matching
  - Kekulization
- Graph traversal
Maximum common substructure (MCS)
- By clique detection algorithm
- Node-induced (MCIS) and edge-induced (MCES)
- Connected and disconnected
- Topological constraint (known as tdMCS)
- Diameter restriction (MCS-DR) and graph-based local similarity (GLS)

MIT license
See Assets/README.md for details of external data sets and their licenses.