GithubHelp home page GithubHelp logo

yjy616 / rxn4chemistry Goto Github PK

View Code? Open in Web Editor NEW

This project forked from rxn4chemistry/rxn4chemistry

0.0 0.0 0.0 6.49 MB

Python wrapper for the IBM RXN for Chemistry API

License: MIT License

Makefile 1.33% Python 97.50% Shell 0.22% Dockerfile 0.95%

rxn4chemistry's Introduction

Python wrapper for the IBM RXN for Chemistry API

Build Status PyPI version License: MIT Binder

logo

A python wrapper to access the API of the IBM RXN for Chemistry website.

Install

From PYPI:

pip install rxn4chemistry

Or directly from the repo:

pip install git+https://github.com/rxn4chemistry/rxn4chemistry.git

Usage

By default, the wrapper connects to the https://rxn.res.ibm.com server. This can be overriden by setting an environment variable. To set a different url, simply do:

export RXN4CHEMISTRY_BASE_URL="https://some.other.rxn.server"

The base url can be directly set when instantiating the RXN4ChemistryWrapper (this will overwrite the environment variable):

api_key = 'API_KEY'
from rxn4chemistry import RXN4ChemistryWrapper

rxn4chemistry_wrapper = RXN4ChemistryWrapper(api_key=api_key, base_url='https://some.other.rxn.server')
# or set it afterwards
# rxn4chemistry_wrapper = RXN4ChemistryWrapper(api_key=api_key)
# rxn4chemistry_wrapper.set_base_url('https://some.other.rxn.server')

Create a project

Get your API key from here and build the wrapper:

api_key = 'API_KEY'
from rxn4chemistry import RXN4ChemistryWrapper

rxn4chemistry_wrapper = RXN4ChemistryWrapper(api_key=api_key)
# NOTE: you can create a project or set an esiting one using:
# rxn4chemistry_wrapper.set_project('PROJECT_ID')
rxn4chemistry_wrapper.create_project('test_wrapper')
print(rxn4chemistry_wrapper.project_id)

Reaction prediction

Running a reaction prediction is as simple as:

response = rxn4chemistry_wrapper.predict_reaction(
    'BrBr.c1ccc2cc3ccccc3cc2c1'
)
results = rxn4chemistry_wrapper.get_predict_reaction_results(
    response['prediction_id']
)
print(results['response']['payload']['attempts'][0]['smiles'])

Extracting actions from a paragraph describing a recipe

Extract the actions from a recipe:

results = rxn4chemistry_wrapper.paragraph_to_actions(
    'To a stirred solution of '
    '7-(difluoromethylsulfonyl)-4-fluoro-indan-1-one (110 mg, '
    '0.42 mmol) in methanol (4 mL) was added sodium borohydride '
    '(24 mg, 0.62 mmol). The reaction mixture was stirred at '
    'ambient temperature for 1 hour.'
)
print(results['actions'])

Retrosynthesis prediction

Predict a retrosynthetic pathway given a product:

response = rxn4chemistry_wrapper.predict_automatic_retrosynthesis(
    'Brc1c2ccccc2c(Br)c2ccccc12'
)
results = rxn4chemistry_wrapper.get_predict_automatic_retrosynthesis_results(
    response['prediction_id']
)
print(results['status'])
# NOTE: upon 'SUCCESS' you can inspect the predicted retrosynthetic paths.
print(results['retrosynthetic_paths'][0])

See here for a more comprehensive example.

Create a synthesis and start it on Robot (or Simulator)

Create a synthesis from a retrosynthesis sequence:

# Each retrosynthetic path predicted has a unique sequence_id that can
# be used to create a new synthesis
response = rxn4chemistry_wrapper.create_synthesis_from_sequence(
    sequence_id=results['retrosynthetic_paths'][0]['sequenceId']
)
print(response['synthesis_id'])

# get the entire list of actions for the entire synthesis, as well as a tree representation
synthesis_tree, ordered_tree_nodes, ordered_list_of_actions = rxn4chemistry_wrapper.get_synthesis_plan(
    synthesis_id=response['synthesis_id']
)
for action in ordered_list_of_actions:
    print(action)

synthesis_status_result = rxn4chemistry_wrapper.start_synthesis(
    synthesis_id=response['synthesis_id']
)
print(synthesis_status_result['status'])

synthesis_status_result = rxn4chemistry_wrapper.get_synthesis_status(
    synthesis_id=response['synthesis_id']
)
print(synthesis_status_result['status'])

# NOTE: upon 'SUCCESS' you can download any analysis reports that are available as pdf
# The pdf contents are returned as text
analysis_actions = rxn4chemistry_wrapper.get_synthesis_actions_with_spectrometer_pdf(
    synthesis_id=response['synthesis_id']
)
pdfs = []
for action in analysis_actions:
    pdfs.append(
        rxn4chemistry_wrapper.get_synthesis_analysis_report_pdf(
            **action
        )
    )
print(pdfs[0])

Forward prediction in batch

It is possible to run a batch of forward reaction predictions without linking them to a project:

response = rxn4chemistry_wrapper.predict_reaction_batch(precursors_list=['BrBr.c1ccc2cc3ccccc3cc2c1', 'Cl.c1ccc2cc3ccccc3cc2c1']*5)
# wait for the predictions to complete
time.sleep(2)
print(rxn4chemistry_wrapper.get_predict_reaction_batch_results(response["task_id"]))

NOTE: the results for batch prediction are not stored permanently in our databases, so we strongly recommend to save them since they will expire.

Examples

To learn more see the examples.

Documentation

The documentation is hosted here using GitHub pages.

rxn4chemistry's People

Contributors

asobczyk avatar drugilsberg avatar ltalirz avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.