https://github.com/pfizer-opensource/torsional-strain

https://github.com/pfizer-opensource/torsional-strain

Last synced: 13 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/pfizer-opensource/torsional-strain
• Owner: pfizer-opensource
• License: apache-2.0
• Created: 2019-01-04T19:05:37.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2023-03-14T13:19:09.000Z (almost 2 years ago)
• Last Synced: 2024-09-26T02:01:40.937Z (4 months ago)
• Language: Python
• Size: 166 KB
• Stars: 20
• Watchers: 6
• Forks: 7
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• top-pharma50 - **pfizer-opensource/torsional-strain** - 03-14 13:19:09 | (Ranked by starred repositories)
• top-pharma50 - **pfizer-opensource/torsional-strain** - 03-14 13:19:09 | (Ranked by starred repositories)

README

        
# Calculation of Torsional Strain Energy

## Setting up the Environment

This step will get you setup with an evironment within which you can run the 

code in this repository using Orion.

1. Get access to OpenEye Orion and Magpie. If you are a licensed Orion user and 

   don't have access, please contact [OpenEye Support]([email protected]).

2. Activate a conda distribution (typically something like `source $CONDA/activate`)

3. Create an environment

        `conda env create -v -f environment.yaml -n torsion`

4. Activate the environment

        `conda activate torsion`

5. Install OpenEye packages  

    a. Insert your magpie token into the first line of `requirements_dev.txt`:  

            `--extra-index-url https://token:[email protected]/simple/`  

    b. Install packages with `pip install -r requirements_dev.txt`

6. Configure the orion CLI  

            `ocli --profile=default config profile`

You should now have an environment capable of running the orion code.

## Uploading the Floe to Orion

1. Activate the environment

2. Run `invoke clean package upload`

This will add a package called `torsional-strain 1.0.0` with a floe called `Torsional Strain Energy` to Orion. 

![Floe Screenshot](floe_screenshot.jpg?raw=true "Floe Screenshot")

## Running the Floe

The floe requires only one input: a dataset of molecules for which we want to calculate torsional strain.  

The floe generates two datasets as primary outputs:

1. `output`: one output molecule per input molecule, with total molecular strain, strains at every torsion, and the torsional energy profiles of all the torsions attached as SD properties

2. `qm_confs`: QM-optimized conformers for each torsional fragment for each molecule  

  

The floe also generates datasets for torsional fragments, pre-QM torsional conformers, and conformers that fail various QM stages.