Skip to content
This repository


Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP

Ruby interface to the OpenBabel ruby bindings similar to pybel

branch: master

Fetching latest commit…


Cannot retrieve the latest commit at this time

Octocat-spinner-32 lib
Octocat-spinner-32 reference
Octocat-spinner-32 spec
Octocat-spinner-32 .document
Octocat-spinner-32 .gitignore
Octocat-spinner-32 .rspec
Octocat-spinner-32 Gemfile
Octocat-spinner-32 LICENSE
Octocat-spinner-32 Rakefile
Octocat-spinner-32 rubabel.gemspec


Ruby interface to the OpenBabel ruby bindings (or the openbabel gem). The interface is ruby-ish analogue of pybel.


  • Smith R, Williamson R, Ventura D, Prince JT. Rubabel: wrapping open Babel with Ruby. J Cheminform. 2013. 5(1):35. pubmed, article (open access)


The Chemistry Toolkit Rosetta Wiki has a lot of examples you can check out.

Creating a molecule

From a string

require 'rubabel'

# by default, reads in smiles strings
serine = Rubabel["C(C(C(=O)O)N)O"]
# more formally:
serine = Rubabel::Molecule.from_string("C(C(C(=O)O)N)O")

# also any other format openbabel supports, for example inchi
serine = Rubabel["InChI=1S/C3H7NO3/c4-2(1-5)3(6)7/h2,5H,1,4H2,(H,6,7)", :inchi]

# from an InChI Key or other id (such as LipidMaps ID) [requires internet]
serine = Rubabel["MTCFGRXMJLQNBG-REOHCLBHSA-N", :inchikey]

Find out all the formats Rubabel supports (hash is format key pointing to the description):

hash = Rubabel.in_formats
hash = Rubabel.out_formats
hash = Rubabel.id_formats # molecule retrieved by web-lookup

From a file

Reading multiple entries from a file:

Rubabel.foreach("file.sdf") do |mol|
  puts mol.exact_mass

Foreach returns an enumerator that can be chained:

# return an array of every unique atom type in the file
uniq_atom_types = Rubabel.foreach("file.mol").flat_map {|mol| }.uniq

Read a single molecule from a file (reads only the first molecule)

mol = Rubabel::Molecule.from_file("file.sdf")
# handles gzipped files seamlessly:
mol = Rubabel::Molecule.from_file("file.sdf.gz") 
mol = Rubabel.molecule_from_file("file.sdf") # alternative

# explicit format for uninformative/wrong extensions:
mol = Rubabel::Molecule.from_file("file", :sdf)


create string output

mol = Rubabel["OCC"] # ethanol

mol.to_s       # canonical smiles -> "CCO"
mol.csmiles    # same thing

mol.to_s(:smi) # smiles -> "OCC"
mol.smiles     # same thing

For inclusion in a file with standard smiles formatting (SMILES\tID\n):

can_smiles_string = mol.write_string # -> "CCO\t\n"
mol.title = "ethanol"
can_smiles_string = mol.write(:can)  # -> "CCO\tethanol\n"

Other formats in the same manner:

pdb_string = mol.write(:pdb)

Write to a file directly (single molecule only; depends on file extension for type):

# write to a smiles file

Write multiple molecules to a file:"somefile.pdb", 'w') do |out|
  molecules.each {|mol| out.print mol.write(:pdb) }


If you write to svg or png (png uses mini_magick to convert from svg) then the molecule is automatically drawn in 2D:

mol = Rubabel["NCC(O)C(=O)O"]

# must have imagemagick ('convert' command) and mini_magick gem installed

Searching and Splitting

each_match, matches, matches?, smarts_indices all take the same input (SMARTS string or object and optional boolean specifying uniqueness of results):

mol = Rubabel["NCC(O)C(=O)O"]
mol.each_match("CO") do |match|
  # match is just an array of atoms that matched
  match.first.el # => :C
  match.last.el # => :O

# matches returns all the matches in an array
all_matches = mol.matches("CO")
# all the match routines take a boolean to alter uniqueness
all_matches = mol.matches("CO", false) # some matches may not be uniq

Have some bonds to break?, split makes new molecules split from that bond(s)

bonds = mol.matches("CO").map {|c, o| c.get_bond(o) }
mol.split(*bonds)  # splits between every carbon single bonded to oxygen

Add, delete, modify atoms/bonds


mol = Rubabel["OCC"]
# adds a carbon, then an oxygen to the previous carbon
mol << 6 << 8    # #<Mol "OCCCO">
mol << :C << :O  # same thing

# add an ethyl group specifically to second atom (the first carbon)
mol = Rubabel["OCC"]
mol[1] << :C << :C

# add a vinyl group to second carbon (use method notation and parenthesis 
# because we are going to specify 2 arguments (the bond order):
( mol[1] << :C).<<(:C, 2)


# delete an atom:
mol = Rubabel["NCO"]
# mol.to_s -> #<Mol CO>

# delete a bond:
bond = mol[0].get_bond(mol[1])
# mol.to_s ->  #<Mol C.O>


Can easily change the bond order:

mol = Rubabel["CC"]
bond = mol[0].get_bond(mol[1])  # get the bond you want
bond = mol[0].bonds.first       # alternate way to get at bonds

bond += 2    # now it is a triple bond
bond.dec!    # now a double bond
bond -= 1    # now a single bond!(2) # back to a triple bond


Should work with ruby 1.9 and 2.0. Many thanks to Andreas Maunz for packaging openbabel as a gem which makes this install quite painless.

Quick Install

On a POSIX system, make sure you have openbabel (including header files), cmake, curl, tar, sed and make {see openbabel instructions}[]. On a recent ubuntu/debian system, this should work

sudo apt-get install ruby ruby-dev openbabel libopenbabel-dev cmake make curl build-essential libcairo-dev

Then install the gem (which should install the openbabel gem, too):

sudo gem install rubabel

Building from Source

  1. download openbabel
  2. swap out Init_OpenBabel for Init_openbabel in scripts/ruby/openbabel-ruby.cpp (see here). Some versions have this fixed already, apparently.
  3. make sure you have the right dependencies to compile

Here's a complete example of compiling for a single user on Ubuntu 11.10 and probably will be generally forward compatible for some time. This will compile bindings on whichever ruby comes up with '/usr/bin/env ruby':

# install the dependencies:
sudo apt-get install libeigen2-dev cmake libwxgtk2.8-dev libxml2-dev libcairo2-dev
# unpack it:
tar -xzvf openbabel-2.3.1.tar.gz
# swap out buggy lines in ruby bindings:
sed -i 's/Init_OpenBabel/Init_openbabel/g' openbabel-2.3.1/scripts/ruby/openbabel-ruby.cpp
# make a separate build directory for building in:
mkdir build-rvmruby1.9.3
cd build-rvmruby1.9.3
mkdir ~/tools
cmake ../openbabel-2.3.1 -DRUBY_BINDINGS=ON -DCMAKE_INSTALL_PREFIX=~/tools/openbabel-rvmruby1.9.3
make && make install

See also


MIT License. See LICENSE for further details.

Something went wrong with that request. Please try again.