Welcome to SyGMa’s documentation!

Contents:

API

class scenario.Rule(rulename, probability, smarts)

Class to contain a metabolic rule

Parameters:
  • rulename – A string containing a unique name of the rule
  • probability – A probability value between 0 and 1 indicating the empirical success rate of the rule
  • smarts – A reaction smarts describing the chemical transformation of the rule
class scenario.Scenario(scenario)

Class to read and process metabolic scenario

Parameters:scenario – A list of lists, each representing a metabolic phase as [name_of_file_containing_rules, number_of_cycles_to_apply]
run(parentmol)
Parameters:parentmol – An RDKit molecule
Returns:A sygma.Tree object
class tree.Tree(parentmol=None)

Class to build and analyse a metabolic tree

Parameters:parentmol – An RDKit molecule
add_coordinates()

Add missing atomic coordinates to all metabolites

calc_scores()

Calculate probability scores for all metabolites

metabolize_all_nodes(rules, cycles=1)

Metabolize all nodes according to [rules], for [cycles] number of cycles

Parameters:
  • rules – List of rules
  • cycles – Integer indicating the number of subsequent steps to apply the rules
to_list(filter_small_fragments=True, parent_column='parent')

Generate a list of metabolites

Parameters:
  • filter_small_fragments – Boolean to activate filtering all metabolites with less then 15% of original atoms (of the parent)
  • parent_column – String containing the name for the column with the parent molecule
Returns:

A list of dictionaries for each metabolites, containing the SyGMa_metabolite (an RDKit Molecule), SyGMa_pathway and SyGMa_score, sorted by decreasing probability.
to_smiles(filter_small_fragments=True)

Generate a smiles list of metabolites

Parameters:filter_small_fragments – Boolean to activate filtering all metabolites with less then 15% of original atoms (of the parent)
Returns:A list of metabolites as list [[SyGMa_metabolite as smiles, SyGMa_score]] sorted by decreasing probability score.
write_sdf(file=<open file '<stdout>', mode 'w'>, filter_small_fragments=True)

Generate an SDFile with metabolites including the SyGMa_pathway and the SyGMa score as properties

Parameters:
  • file – The SDF file to write to
  • filter_small_fragments – Boolean to activate filtering all metabolites with less then 15% of original atoms (of the parent)
class treenode.TreeNode(mol, parent='', rule=None, score=None, pathway='')

Class containing a node of the SyGMa tree

Key mol:RDKit Molecule
Key parents:Dictonary {inchikey_of_parent: rulename_transforming_parent_to_self}
Key children:List of inchikeys of the child nodes
Key score:Value between 0 and 1
Key pathway:String describing the pathway from parent to self
Key n_original_atoms:
 Integer, number of atoms originating from parent or None if not yet determined
gen_coords()

Calculate 2D positions for atoms in self.mol without coordinates

Command line script

SyGMa: Systematically Generating potential Metabolites

usage: sygma [-h] [--version] [-o OUTPUTTYPE] [-1 PHASE1] [-2 PHASE2]
             [-l {debug,info,warn, error}]
             parentmol

Positional Arguments

parentmol Smiles string of parent molecule structure

Named Arguments

--version show program’s version number and exit
-o, --outputtype
 

Molecule output type (default: sdf)

Default: sdf

-1, --phase1

Number of phase 1 cycles (default: 1)

Default: 1

-2, --phase2

Number of phase 2 cycles (default: 1)

Default: 1

-l, --loglevel

Possible choices: debug, info, warn, error

Set logging level (default: “info”)

Default: “info”

Introduction

SyGMa is a python library for the Systematic Generation of potential Metabolites. It is a reimplementation of the metabolic rules outlined in Ridder, L., & Wagener, M. (2008) SyGMa: combining expert knowledge and empirical scoring in the prediction of metabolites. ChemMedChem, 3(5), 821-832.

Requirements

SyGMa requires RDKit with INCHI support

Installation

  • Install with Anaconda: conda install -c 3d-e-Chem -c rdkit sygma

OR

AND

  • pip install sygma OR, after downloading sygma, python setup.py install

Example

import sygma
from rdkit import Chem


def test_predict_phenol_metabolites():
    """Test prediction of phenol metabolites by sygma module"""

    # Each step in a scenario lists the ruleset and the number of reaction cycles to be applied
    scenario = sygma.Scenario([
        [sygma.ruleset['phase1'], 1],
        [sygma.ruleset['phase2'], 1]])

    # An rdkit molecule, optionally with 2D coordinates, is required as parent molecule
    parent = Chem.MolFromSmiles("c1ccccc1O")

    metabolic_tree = scenario.run(parent)
    metabolic_tree.calc_scores()

    metabolite_list = metabolic_tree.to_list()
    assert len(metabolite_list) == 12
    assert metabolite_list[0]['SyGMa_score'] == 1
    assert metabolite_list[1]['SyGMa_pathway'] == 'O-glucuronidation_(aromatic_hydroxyl); \n'

Docker

SyGMa can be executed in a Docker container as follows:

docker run 3dechem/sygma c1ccccc1O

Rulesets

SyGMa comes currently with two rulesets:

phase1
Phase 1 metabolism rules include mainly different types of oxidation, hydrolysis, reduction and condensation reactions
phase2
Phase 2 metabolism rules include severaly conjugation reaction, i.e. with glucuronyl, sulfate, methyl and acetyl

Indices and tables