Supplementary figures, tables, and code used to train models and produce input files for "Predicting fungal secondary metabolite activity from biosynthetic gene cluster data using machine learning" (Riedling et al. 2023).

All biosynthetic gene cluster and secondary metabolite bioactivities used to train the models can be found in supplemental tables 1 and 2. The Jupyter notebooks fungal_only_trained_model and NP_prediction_bacterial_and_fungal contain the code necessary to train the models. The Jupyter notebooks titled Running_predictions contain the code to make predictions on SM bioactivity from antiSMASH and RGI input. The Jupyter notebooks require the additional Python scripts so keep all scripts in the same directory as the notebooks. The scripts to extract the features from the antiSMASH and RGI files are included in the notebooks.

用于训练模型的全部生物合成基因簇（biosynthetic gene cluster）及次级代谢产物（secondary metabolite）生物活性数据，均可在补充表1与补充表2中获取。用于训练模型的必要代码收录于名为fungal_only_trained_model和NP_prediction_bacterial_and_fungal的Jupyter笔记本（Jupyter Notebook）中。名为Running_predictions的Jupyter笔记本则包含了基于antiSMASH（antiSMASH）与RGI（RGI）输入文件进行次级代谢产物生物活性预测的代码。运行该类Jupyter笔记本需依赖额外的Python脚本，请将所有脚本与笔记本文件置于同一目录下。从antiSMASH与RGI文件中提取特征的脚本已集成于前述Jupyter笔记本中。