scTopoGAN

scTopoGAN is a method for unsupervised manifold alignment of single-cell datasets with non-overlapping cells or features.

The method uses topological autoencoders (topoAE) to obtain latent representations of each modality separately. A topology-guided Generative Adversarial Network then aligns these latent representations into a common space.

For details and the software, please visit the scTopoGAN GitHub page and read publication by Singh et al. (2023).

scTopoGAN overview. scTopoGAN consists of two stages: (1) Manifold projection using topological autoencoders to obtain a low-dimensional embedding (manifold) for each modality independently. (2) Manifold alignment using a GAN.