ADDZYME is a machine learning-based algorithm for the prediction of the effect of an additive on the activity of an enzyme. It uses 30 ERT-Baseline models to report relative activity. For more information refer to our published paper.
Milad Rayka, [email protected]
ADDZYME: A software to predict effect of additives on enzyme activity.
Below packages should be installed for using ADDZYME. Dependencies:
-
python = 3.8.16
-
numpy = 1.23.5
-
streamlit = 1.25.0
-
molvs = 0.1.1
-
joblib = 1.2.0
-
rdkit-pypi (2022/9/5 release)
-
scikit-learn = 1.2.2
For installing first make a virtual environment and activate it.
On windows:
python py -m venv env .\env\Scripts\activate
On macOS and Linux:
python3 -m venv env source env/bin/activate
Which env is the location to create the virtual environment. Now, you can install packages:
pip install package_name==*version"
For using ADDZYME, after activating your environment, e.g., env, change your directory to ./codes and type the following command:
streamlit run addzyme_gui.py
For reproducing all results of the paper use preparation_and_training.ipynb and dataset_analysis.ipynb. Gathered data points are available in enzyme_additive_dataset.csv. Check the codes folder.