NAME

Chemistry::File::SMILES - SMILES linear notation parser/writer

SYNOPSYS

    #!/usr/bin/perl
    use Chemistry::File::SMILES;

    # parse a SMILES string
    my $s = 'C1CC1(=O)[O-]';
    my $mol = Chemistry::Mol->parse($s, format => 'smiles');

    # print a SMILES string
    print $mol->print(format => 'smiles');

    # print a unique (canonical) SMILES string
    print $mol->print(format => 'smiles', unique => 1);

    # parse a SMILES file
    my @mols = Chemistry::Mol->read("file.smi", format => 'smiles');

    # write a multiline SMILES file
    Chemistry::Mol->write("file.smi", mols => \@mols);

DESCRIPTION

This module parses a SMILES (Simplified Molecular Input Line Entry Specification) string. This is a File I/O driver for the PerlMol project. <http://www.perlmol.org/>. It registers the 'smiles' format with Chemistry::Mol.

This parser interprets anything after whitespace as the molecule's name; for example, when the following SMILES string is parsed, $mol->name will be set to "Methyl chloride":

    CCl  Methyl chloride

The name is not included by default on output. However, if the "name" option is defined, the name will be included after the SMILES string, separated by a tab.

    print $mol->print(format => 'smiles', name => 1);

Multiline SMILES and SMILES files

A file or string can contain multiple molecules, one per line.

    CCl  Methyl chloride
    CO   Methanol

Files with the extension '.smi' are assumed to have this format.

Atom Mapping Numbers

As an extension for reaction processing, SMILES strings may have atom mapping numbers, which are introduced after a colon in a bracketed atom. For example, [C:1]. The mapping number need not be unique. This module reads the mapping numbers and stores them as the name of the atom ($atom->name).

On output, atom names are not included by default. See the "number" and "auto_number" options below for ways of including them.

head1 OPTIONS

The following options are supported in addition to the options mentioned for Chemistry::File, such as "mol_class", "format", and "fatal".

aromatic

On output, detect aromatic atoms and bonds by means of the Chemistry::Ring module, and represent the organic aromatic atoms with lowercase symbols.

unique

When used on output, canonicalize the structure if it hasn't been canonicalized already and generate a unique SMILES string. This option implies "aromatic".

number

For atoms that have a defined name, print the name as the "atom number". For example, if an ethanol molecule has the name "42" for the oxygen atom and the other atoms have undefined names, the output would be:

    CC[OH:42]

auto_number

When used on output, number all the atoms explicitly and sequentially. The output for ethanol would look something like this:

    [CH3:1][CH2:2][OH:3]

name

Include the molecule name on output, as described in the previous section.

kekulize

When used on input, assign single or double bond orders to "aromatic" or otherwise unspecified bonds (i.e., generate the Kekule structure). If false, the bond orders will remain single. This option is true by default. This uses "assign_bond_orders" from the Chemistry::Bond::Find module.

CAVEATS

Stereochemistry is not supported! Stereochemical descriptors such as @, @@, /, and \ will be silently ignored on input, and will certainly not be produced on output.

Reading branches that start before an atom, such as (OC)C, which should be equivalent to C(OC) and COC, according to some variants of the SMILES specification. Many other tools don't implement this rule either.

The kekulize option works by increasing the bond orders of atoms that don't have their usual valences satisfied. This may cause problems if you have atoms with explicitly low hydrogen counts.

VERSION

0.47

AUTHOR

Ivan Tubert-Brohman <itub@cpan.org>