How to Build the Virtual Cell with Artificial Intelligence

Now that the article is out and Open Access on Cell, I am reposting a comment I made a few months ago about a super interesting preprint, "How to Build the Virtual Cell with Artificial Intelligence: Priorities and Opportunities" (list of authors below), which describes one of the most exciting opportunities in biology, the next frontier in a way -one we thought we understood but rediscover continually: the cell.

The reductionist view that the genome is the answer to everything (ok, a bit of a caricature but you get the point) is being replaced slowly by the incredible complexity of what happens inside the cell. The emergence of new technologies, such as single cell sequencing and imaging techniques, combined with a phase transition in AI and modeling performance, means that we can start building a virtual cell, that is, a "realistic" model of the cell, where the level of realism gets progressively deeper and deeper. We just have to remember that a model is an understandable, explainable, controllable, identifiable, simplified representation of reality designed to address a particular question. For example, a spatial model can help address how the location of proteins inside the various cell compartments impact cell function, or how many ribosomal proteins can be expressed simultaneously.

It's just an area of incredible potential. Thank you!