Chrfinder 0.0.2

The package find the best chromatography based on properties of the mixture

Chrfinder

Project overview

Welcome to Chrfinder! This project automates the selection of the most suitable chromatography technique . By simply providing the names of the molecules in the mixture, the code retrieves their physicochemical properties from PubChem (web source) and determines the optimal chromatography method based on these properties. It also gives the optimal conditions.

✅ Benefits

• 🚀 Efficiency: Automates the property retrieval and decision-making process, saving time and reducing manual effort.
• 🎯 Accuracy: Utilizes precise physicochemical data to ensure the most suitable chromatography technique is chosen.
• 🌐 Versatility: Supports a wide range of organic compounds and chromatography methods (PubChem database).

⚙ Installation

Create a new environment, you may also give the environment a different name.

conda create -n chrfinder python=3.10 

conda activate chrfinder
pip install Chrfinder

🛠️ Installation for Development

To get started with Chrfinder, you can follow these steps:

git clone https://github.com/Averhv/Chrfinder.git
cd Chrfinder
pip install .
pip install jupyterlab
jupyter lab

Optional: Installing for Development and Testing

If you want to contribute to Chrfinder or run the tests and get coverage, you can install the package in editable mode along with the necessary dependencies for testing and documentation:

git clone https://github.com/Averhv/Chrfinder.git
cd Chrfinder
pip install -e ".[test,doc]"
pip install jupyterlab
jupyter lab

Then you need to run the tests as follow in your terminal:

pip install tox
tox

One can also run the following to get better readability:

• runs tests for the Chrfinder.py file
• measures code coverage
• generates coverage reports in both XML and terminal formats
• provides verbose output during test execution.

pytest --cov=src/Chrfinder.py --cov-report xml:.tox/coverage.xml --cov-report term -vv

Test result: 15 passed in ~20s

📒 Features

from chrfinder import main

# Running the whole file ask for molecules through Tkinter and returns the best chromatography
main()

🌐 Optional functions

find_pka(inchikey)

Finds the pKa value for a compound using its InChIKey.

from chrfinder import find_pka

inchikey = "XEFQLINVKFYRCS-UHFFFAOYSA-N"
find_pka(inchikey)

find_boiling_point(name)

Finds the boiling point for a compound by name.

from chrfinder import find_boiling_point

compound_name = "Ethanol"
find_boiling_point(compound_name)

get_df_properties()

Get a DataFrame of properties for a mixture of compounds.

from chrfinder import get_df_properties

mixture = ["Acetone", "Ethanol", "Methanol"]
get_df_properties(mixture, verbose=True)

How It Works

1. Input: User provides the names of the molecules present in the mixture through a Tkinter interface.

2. Data Retrieval: Finds the following key physicochemical properties for each molecule in Pubchem:

   • Boiling temperature (°C)
   • logP (partition coefficient)
   • pKa (acid dissociation constants)
   • Molecular mass

3. Chromatography Type Decision: Follows logical conditions to determine best chromatography and conditions

   • Gas Chromatography (GC): if the Boiling Point is low (T_eb <250°C).
   • Ion Chromatography (IC): for small molecules (M<2000g/mol) and a negative maximum LogP negative
     • Selected if the maximum molecular mass is less than or equal to 2000, and the maximum logP is negative, with a proposed pH derived from the pKa values.
   • High-Performance Liquid Chromatography (HPLC): Chosen for different conditions. Stationary phases and eluent natures are suggested.
   • Size Exclusion Chromatography (SEC): For big molecules (M>2000g/mol). From LogP, it suggest gel permeation or gel filtration, with corresponding eluant.

4. Output:

   • The code outputs the advisable chromatography type, the nature of the eluent (gas, aqueous, or organic), and the proposed pH for the eluent if applicable through the Tkinter interface.

Work in progress...

• Build a data molecules thermostability database;
• taking into account multiple pKa values for polyacids for exemple;
• optimize the research: search only one time te same name;
• find physicalchemical properties as addition functionality;

🫱🏽‍🫲🏼 Contributing

Contributions are welcome! Please submit a pull request or open an issue to discuss any changes.