Published October 6, 2022 | Version 1.0
Software Open

T-GAN-D: a GAN-based classifier for breast cancer prognostication

  • 1. Institute of Cell Biology and Immunology, University of Stuttgart
  • 2. Prokando GmbH
  • 3. Institute of Cell Biology and Immunology, Stuttgart Research Center Systems Biology, University of Stuttgart

Description

T-GAN-D is a generative adversarial networks-base classifier for breast cancer prognostication.
Full transcriptome profiles are used as input to discriminate high vs. low risk patients as described in the original article: "Applying GAN-based data augmentation to improve transcriptome-based prognostication in breast cancer".

The archive contains the R script for downloading, rescaling and merging the METABRIC and MRCA-TCGA cohorts used as use cases in the study. Input files (clinical and expression data) used to train and test the T-GAN-D as described in the original article are provided, together with the python scripts used for data pre-processing and generation of risk class predictions.

The study used openly available human data that were originally available through cBioportal.org, the R package MetaGxBreast (Gendoo et al. 2019, DOI: 10.1038/s41598-019-45165-4) and the following publications: Xia et al. 2019 (DOI: 10.1038/s41467-019-13588-2); Rueda et al. 2019 (DOI: 10.1038/s41586-019-1007-8); Liu et al. 2018 (DOI: 10.1016/j.cell.2018.02.052).

Notes

MR and CG receive funding by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy - EXC 2075 – 390740016 and acknowledge the support by the Stuttgart Center for Simulation Science (SimTech).

Files

Files (628.3 MB)

Name Size Download all
md5:736e3c598b86eb31a4b1159bb8d81c4a
628.3 MB Download