Skip to content

Machine Learning Benchmark for Transferase Substrate Activation: This repository presents a systematic approach to evaluate a comprehensive set of features for predicting activation free energy, specifically focusing on kcat values. The benchmark model automates the process, ensuring rigorous and reproducible results in substrate energy screening.

License

bryankappa/TransferaseML-Benchmark

Repository files navigation

Introduction

A machine learning based algorithms benchmark for the reductase substrate spectrum activation free energy energy screening. This is a benchmark that consist of SVM, RandomForest, XGboost, and neural network. A algorithm metric for measuring Kcat value. image

image

Installing dependencies

pip install RDKit scikit-learn selfies numpy pandas

Running the Machine learning pipeline

run in command line and change the pipeline that is needed.

python main.py

Results

Amino Acid one hot encoding Tranferase. image

About

Machine Learning Benchmark for Transferase Substrate Activation: This repository presents a systematic approach to evaluate a comprehensive set of features for predicting activation free energy, specifically focusing on kcat values. The benchmark model automates the process, ensuring rigorous and reproducible results in substrate energy screening.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published