This directory contains instructions for reproducing the results in our paper:
Swanson, K., Liu, G., Catacutan, D. B., Arnold, A., Zou, J., Stokes, J. M. Generative AI for designing and validating easily synthesizable and structurally novel antibiotics. Nature Machine Intelligence, 2024.
The relevant data should be downloaded from this Zenodo record, unzipped, and moved to SyntheMol/data.
Note that here, we use the 2021 q3-4 version of the building blocks and the 2022 q1-2 version of the enumerated REAL Space molecules. However, by default, SyntheMol now downloads a newer version of the building blocks during installation.
real.md: Instructions for processing Enamine REAL building blocks, reactions, and molecules.
clogp.md: Instructions for performing an in silico study of SyntheMol using a computational molecular property, cLogP, which is the computed octanol-water partition coefficient.
antibiotics.md: Instructions for generating antibiotic candidates for Acinetobacter baumannii. Includes instructions for processing antibiotics data, training antibacterial activity prediction models, generating molecules with SyntheMol, and selecting candidates.
plots.md Instructions for producing plots analyzing the data and results.