Researchers have developed -- scGen -- an Artificial Intelligence (AI)-powered tool which promises to reshape the way we study diseases and their treatment at cellular level. The study, published in the journal Nature Methods, shows scGen will help mapping and studying cellular response to diseases and their treatment beyond experimentally available data.
According to the researchers, scGen is a generative deep learning model that leverages ideas from image, sequence and language processing and applies them to model the behaviour of a cell performed on computer or via computer simulation.
The next step for the team will be to improve scGen, make it a fully data-driven formulation, increasing its predictive power to enable the study of combinations of perturbations.
"We can now start optimising scGen to answer more and more complex questions about diseases," said the researcher Alex Wolf from the Technical University of Munich in Germany.
Large-scale atlases of organs in a healthy state are soon going to be available, in particular, within the Human Cell Atlas. This is a significant step in understanding cells, tissues and organs in healthy state in a better way and providing a reference while diagnosing, monitoring and treating diseases.
Accurately modelling cellular response to perturbations e.g. disease, compounds and genetic interventions, is a central goal of computational biology.
In addition, scGen is the first tool that predicts cellular response out-of-sample. This means that scGen, if trained on data that captures the effect of perturbations for a given system, is able to make reliable predictions for a different system.
"For the first time, we have the opportunity to use data generated in one model system such as mouse and use the data to predict disease or therapy response in human patients," said Mohammad Lotfollahi from Technical University of Munich.