Generative adversarial networks uncover epidermal regulators and predict single cell perturbations
Arsham Ghahramani and Fiona M Watt and Nicholas M Luscombe
bioRxiv: The preprint server for biology - 2018 via Local CrossRef

Summary by David Stutz 5 years ago
Your comment:
Summary by arshamg 6 years ago
Is there any objective evaluation that this approach is better than another or a simple baseline? What evaluation would I use to compare against the performance of this model?

Generative models for scRNA-seq are new so there's not currently a way to compare different generative models. There are ways to compare different dimensionality reduction methods, especially if you have a ground-truth mixture of labelled cells. Our main objective measure of performance is the predictions we make (e.g. prediction of genes that change over differentiation in the skin) and how these match up to real skin cell behaviour.

Your comment:

ShortScience.org allows researchers to publish paper summaries that are voted on and ranked!

Sponsored by: