K Kumar Inorganic Chemistry 27.pdf
Bente Coker
A frequent problem in computational modeling is the interconversion of chemical structures between different formats. While standard interchange formats exist (for example, Chemical Markup Language) and de facto standards have arisen (for example, SMILES format), the need to interconvert formats is a continuing problem due to the multitude of different application areas for chemistry data, differences in the data stored by different formats (0D versus 3D, for example), and competition between software along with a lack of vendor-neutral formats.
We discuss, for the first time, Open Babel, an open-source chemical toolbox that speaks the many languages of chemical data. Open Babel version 2.3 interconverts over 110 formats. The need to represent such a wide variety of chemical and molecular data requires a library that implements a wide range of cheminformatics algorithms, from partial charge assignment and aromaticity detection, to bond order perception and canonicalization. We detail the implementation of Open Babel, describe key advances in the 2.3 release, and outline a variety of uses both in terms of software products and scientific research, including applications far beyond simple format interconversion.
Open Babel presents a solution to the proliferation of multiple chemical file formats. In addition, it provides a variety of useful utilities from conformer searching and 2D depiction, to filtering, batch conversion, and substructure and similarity searching. For developers, it can be used as a programming library to handle chemical data in areas such as organic chemistry, drug design, materials science, and computational chemistry. It is freely available under an open-source license from
The history of chemical informatics has included a huge variety of textual and computer representations of molecular data. Such representations focus on specific atomic or molecular information and may not attempt to store all possible chemical data. For example, line notations like Daylight SMILES [1] do not offer coordinate information, while crystallographic or quantum mechanical formats frequently do not store chemical bonding data. Hydrogen atoms are frequently omitted from x-ray crystallography due to the difficulty in establishing coordinates, and are often ignored by some file formats as the "implicit valence" of heavy atoms that indicates their presence. Other types of representations require specification of atom types on the basis of a specific valence bond model, inclusion of computed partial charges, indication of biomolecular residues, or multiple conformations.
We outline for the first time, the development and use of the Open Babel project, a full-featured open chemical toolbox, designed to "speak" the many different representations of chemical data. It allows anyone to search, convert, analyze, or store data from molecular modeling, chemistry, solid-state materials, biochemistry, or related areas. It provides both ready-to-use programs as well as a complete, extensible programmer's toolkit for developing cheminformatics software. It can handle reading, writing, and interconverting over 110 chemical file formats, supports filtering and searching molecule files using Daylight SMARTS pattern matching [7] and other methods, and provides extensible fingerprinting and molecular mechanics frameworks. We will discuss the frameworks for file format interconversion, fingerprinting, fast molecular searching, bond perception and atom typing, canonical numbering of molecular structures and fragments, molecular mechanics force fields, and the extensible interfaces provided by the software library to enable further chemistry software development.
Open Babel has its origin in a version of OELib released as open-source software by OpenEye Scientific under the GPL (GNU Public License). In 2001, OpenEye decided to rewrite OELib in-house as the proprietary OEChem library, so the existing code from OELib was spun out into the new Open Babel project. Since 2001, Open Babel has been developed and substantially extended as an international collaborative project using an open-source development model [8]. It has over 160,000 downloads, over 400 citations [9], is used by over 40 software projects [10], and is freely available from the Open Babel website [11].
With the release of Open Babel 2.3, Open Babel supports 111 chemical file formats in total. It can read 82 formats and write 85 formats. These encompass common formats used in cheminformatics (SMILES, InChI, MOL, MOL2), input and output files from a variety of computational chemistry packages (GAMESS, Gaussian, MOPAC), crystallographic file formats (CIF, ShelX), reaction formats (MDL RXN), file formats used by molecular dynamics and docking packages (AutoDock, Amber), formats used by 2D drawing packages (ChemDraw), 3D viewers (Chem3D, Molden) and chemical kinetics and thermodynamics (ChemKin, Thermo). Formats are implemented as "plugins" in Open Babel, which makes it easy for users to contribute new file formats (see Extensible Interface below). Depending on the format, other data is extracted by Open Babel in addition to the molecular structure; for example, vibrational frequencies are extracted from computational chemistry log files, unit cell information is extracted from CIF files, and property fields are read from SDF files.
A number of "utility" file formats are also defined; these are not strictly speaking a way of storing the molecular structure, but rather present certain functionality through the same interface as the regular file formats. For example, the report format is a write-only utility format [12] that presents a summary of the molecular structure of a molecule; the fingerprint format [13] and fastsearch format [14] are used for similarity and substructure searching (see below); the MolPrint2D and Multilevel Neighborhoods of Atoms formats calculate circular fingerprints defined by Bender et al. [15, 16] and Filimonov et al. [17, 18] respectively.
Each format can have multiple options to control either reading or writing a particular format. For example, the InChI format has 12 options including an option "K" to generate an InChIKey, "T " to truncate the InChI depending on a supplied parameter and "w" to ignore certain InChI warnings. The available options are listed in the documentation, are shown in the Graphical User Interface (GUI) as checkboxes or textboxes, and can be listed at the command-line. In fact, all three are generated from the same source; a documentation string in the C++ code.
Databases are widely used to store chemical information especially in the pharmaceutical industry. A key requirement of such a database is the ability to index chemical structures so that they can be quickly retrieved given a query substructure. Open Babel provides this functionality using a path-based fingerprint. This fingerprint, referred to as FP2 in Open Babel, identifies all linear and ring substructures in the molecule of lengths 1 to 7 (excluding the 1-atom substructures C and N) and maps them onto a bit-string of length 1024 using a hash function. If a query molecule is a substructure of a target molecule, then all of the bits set in the query molecule will also be set in the target molecule. The fingerprints for two molecules can also be used to calculate structural similarity using the Tanimoto coefficient, the number of bits in common divided by the union of the bits set.
Clearly, repeated searching of the same set of molecules will involve repeated use of the same set of fingerprints. To avoid the need to recalculate the fingerprints for a particular multi-molecule file (such as an SDF file), Open Babel provides a fastindex format that solely stores a fingerprint along with an index into the original file. This index leads to a rapid increase in the speed of searching for matches to a query - datasets with several million molecules are easily searched interactively. In this way, a multi-molecule file may be used as a lightweight alternative to a chemical database system.
As mentioned above, many chemical file formats offer representations of molecular data solely as lists of atoms. For example, most quantum chemical software packages and most crystallographic file formats do not offer definitions of bonding. A similar situation occurs in the case of the Protein Data Bank (PDB) format; while standardized [19] files contain connectivity information, non-standard files exist that often do not provide full connectivity information. Consequently, Open Babel features methods to determine bond connectivity, bond order perception, aromaticity determination, and atom typing.
Bond connectivity is determined by the frequently used algorithm of detecting atoms closer than the sum of their covalent radii, with a slight tolerance (0.45 ) to allow for longer than typical bonds. To handle disorder in crystallographic data (e.g., PDB or CIF files), atoms closer than 0.63 are not bonded. A further filtering pass is made to ensure standard bond valency is maintained; each element has a maximum number of bonds, if this is exceeded then the longest bonds to an atom are successively removed until the valence rule is fulfilled.
After bond connectivity is determined, if needed or requested by the user, bond order perception is performed on the basis of bond angles and geometries. The method is similar to that proposed by Roger Sayle [20] and uses the average bond angle around an un-typed atom to determine sp and sp2 hybridized centers. 5-membered and 6-membered rings are checked for planarity to estimate aromaticity. Finally, atoms marked as unsaturated are checked for an unsaturated neighbor to give a double or triple bond. After this initial atom typing, known functional groups are matched, followed by aromatic rings, followed by remaining unsatisfied bonds based on a set of heuristics for short bonds, atomic electronegativity, and ring membership.
b37509886e