BioTransformer is also available as a downloadable, command-line only program which can be accessed here. The installed version allows users to run multiple instances and to perform any number of biotransformation reactions on any number of molecules. It is ideal for large scale processing.

Runnable jar file repository available at: here
Plain java file source code repository available at: here
Program requirements: Single CPU processor with at least 1024 MB RAM and 1 GB free hard disk space.
Operating System: UNIX-based (macOS/Linux)
Installation instructions: These are available here
Operating instructions: These are available here
Hold-out dataset for 68 reactants can be found as the first 68 structures in the EBoMD2.sdf dataset at: here

The downloadable version of BioTransformer is only minimally supported. It is intended for expert users.

Please cite:

Wishart DS, Tian S, Allen D, Oler E, Peters H, Lui VW, Gautam V, Djoumbou Feunang Y, Greiner R, Metz TO; BioTransformer 3.0 – A Web Server for Accurately Predicting Metabolic Transformation Products [Submitted in Nucleic Acids Research, Webserver Issue.Apr.2022]

Djoumbou Feunang Y, Fiamoncini J, de la Fuente AG, Manach C, Greiner R, and Wishart DS; BioTransformer: A Comprehensive Computational Tool for Small Molecule Metabolism Prediction and Metabolite Identification; Journal of Cheminformatics; 2019; Journal of Cheminformatics 11:2; DOI: 10.1186/s13321-018-0324-5