Prediction of bacterial pH preferences using machine learning

This code includes hyperparameter optimization and evaluation of a boosted regression model to predict bacterial pH preferences using information on the presence/absence of functional genes in bacterial genomes. For a more detailed description of the approach see https://github.com/fiererlab/ph_preference/blob/main/Whole_FeatureSelection_Final.ipynb. <br> Important file descriptions: <br> load_and_run_pH_model.ipynb is the notebook required to run the model. pH_model_2022-10-17_56_full.sav is the model file (trained on the data from this study). Readme.txt specifies how to set up an environment with the dependencies found in environment.yaml, which are required to run the notebook and model.