rpreactor

A command-line and python package to handle biochemical reaction rules.

rpreactor is designed to use reaction rules from RetroRules, and to be at the core of more complex bioretrosynthesis tools such as RetroPathRL. It relies extensively on RDKit to handle chemicals and reactions.

Please submit your questions or any new issue you may encounter with rpreactor using GitHub's issue system.

If you are interested in rpreactor you may also be interested in:

• Reactor: ChemAxon's "A high performance virtual synthesis engine"
• ATLAS: "A repository of all possible biochemical reactions for synthetic biology and metabolic engineering studies" by our friends from the LCSB
• MINE: "Metabolic In silico Network Expansion (MINE) Database Construction and DB Logic" by our friends from the Tyo Lab

Installation

Important: rpreactor works with Python >=3.6 and was tested with rdkit 2020.03*.

We strongly recommend you to use conda package manager, and to follow those steps:

# installation in a new conda environment <myenv>
conda create --name <myenv> -c conda-forge -c brsynth rdkit=2020.03 rpreactor
conda activate <myenv>

# installation in an already existing environment <myenv>
conda activate <myenv>
conda install -c conda-forge -c brsynth rdkit=2020.03 rpreactor

Usage

From command line:

conda activate <myenv>
python -m rpreactor.cli --help
python -m rpreactor.cli --with_hs true inline --inchi "InChI=1/C3H6O3/c1-2(4)3(5)6/h2,4H,1H3,(H,5,6)" --rsmarts "([#8&v2:1](-[#6&v4:2](-[#6&v4:3](-[#8&v2:4]-[#1&v1:5])=[#8&v2:6])(-[#6&v4:7](-[#1&v1:8])(-[#1&v1:9])-[#1&v1:10])-[#1&v1:11])-[#1&v1:12])>>([#15&v5](=[#8&v2])(-[#8&v2]-[#1&v1])(-[#8&v2]-[#1&v1])-[#8&v2:1]-[#6&v4:2](-[#6&v4:3](-[#8&v2:4]-[#1&v1:5])=[#8&v2:6])(-[#6&v4:7](-[#1&v1:8])(-[#1&v1:9])-[#1&v1:10])-[#1&v1:11].[#7&v3](=[#6&v4]1:[#7&v3]:[#6&v4](-[#8&v2]-[#1&v1]):[#6&v4]2:[#7&v3]:[#6&v4](-[#1&v1]):[#7&v3](-[#6&v4]3(-[#1&v1])-[#8&v2]-[#6&v4](-[#6&v4](-[#8&v2]-[#15&v5](=[#8&v2])(-[#8&v2]-[#1&v1])-[#8&v2]-[#15&v5](-[#8&v2]-[#1&v1:12])(=[#8&v2])-[#8&v2]-[#1&v1])(-[#1&v1])-[#1&v1])(-[#1&v1])-[#6&v4](-[#8&v2]-[#1&v1])(-[#1&v1])-[#6&v4]-3(-[#8&v2]-[#1&v1])-[#1&v1]):[#6&v4]:2:[#7&v3]:1-[#1&v1])-[#1&v1])"

From within a script:

import json
import rpreactor

inchi = 'InChI=1/C3H6O3/c1-2(4)3(5)6/h2,4H,1H3,(H,5,6)'
rsmarts = '([#8&v2:1](-[#6&v4:2](-[#6&v4:3](-[#8&v2:4]-[#1&v1:5])=[#8&v2:6])(-[#6&v4:7](-[#1&v1:8])(-[#1&v1:9])-[#1&v1:10])-[#1&v1:11])-[#1&v1:12])>>([#15&v5](=[#8&v2])(-[#8&v2]-[#1&v1])(-[#8&v2]-[#1&v1])-[#8&v2:1]-[#6&v4:2](-[#6&v4:3](-[#8&v2:4]-[#1&v1:5])=[#8&v2:6])(-[#6&v4:7](-[#1&v1:8])(-[#1&v1:9])-[#1&v1:10])-[#1&v1:11].[#7&v3](=[#6&v4]1:[#7&v3]:[#6&v4](-[#8&v2]-[#1&v1]):[#6&v4]2:[#7&v3]:[#6&v4](-[#1&v1]):[#7&v3](-[#6&v4]3(-[#1&v1])-[#8&v2]-[#6&v4](-[#6&v4](-[#8&v2]-[#15&v5](=[#8&v2])(-[#8&v2]-[#1&v1])-[#8&v2]-[#15&v5](-[#8&v2]-[#1&v1:12])(=[#8&v2])-[#8&v2]-[#1&v1])(-[#1&v1])-[#1&v1])(-[#1&v1])-[#6&v4](-[#8&v2]-[#1&v1])(-[#1&v1])-[#6&v4]-3(-[#8&v2]-[#1&v1])-[#1&v1]):[#6&v4]:2:[#7&v3]:1-[#1&v1])-[#1&v1])'

o = rpreactor.RuleBurner(rsmarts_list=[rsmarts], inchi_list=[inchi], with_hs=True)
o.compute()
res = json.loads('[' + ', '.join(o._json) + ']')

For developers

Development installation

After a git clone:

cd <repository>
conda env create -f environment.yml -n <dev_env>
conda activate <dev_env>
conda develop -n <dev_env> .

Warning: if you do not specify an environment name with -n <dev_env>, then 'dev_rpreactor' will be used.

Test your installation with:

conda activate <dev_env>
python -m rpreactor.cli -h

To uninstall:

conda deactivate
conda env remove -n <dev_env>

Test suite

cd <repository>
python -m pytest --doctest-modules 

Build the documentation

To build a local Sphinx HTML documentation at <repository>/docs/_build/html/index.html:

cd <repository>/docs
make html

Build and deployment

The process is automated with GitHub's Action.

If you want to check the build process locally:

CONDA_BLD_PATH=<repository>/conda-bld
mkdir -p ${CONDA_BLD_PATH} 
cd <repository>

conda env create -f recipe/conda_build_env.yaml -n <build_env>
conda activate <build_env>
conda build -c conda-forge --output-folder ${CONDA_BLD_PATH} recipe

conda convert --platform osx-64 --platform linux-64 --platform win-64 --output-dir ${CONDA_BLD_PATH} ${CONDA_BLD_PATH}/*/rpreactor-*