With the accumulation of a lot of the transcriptomic data a large amount of expression studies have become centric to biomarkers and their use in different biological state. Treatment predictions are the most common type of data harmonization tasks being carrying out these days. Different style components can be used for carrying out biological details of the samples.
Russkikh N. et. al. (2020) have developed a program to implement such a studies which can be accessed at https://github.com/NRshka/stvae-source.
Reference:
Russkikh N. et. al. (2020) Style transfer with variational autoencoders is a promising approach to RNA-Seq data harmonization and analysis. Bioinformatics 36(20): 5076-5085.