This paper proposes the creation of a Perturbation Cell Atlas to complement the existing Human Cell Atlas, aiming to understand the causal circuits governing human cell and tissue biology. It highlights how recent experimental and computational advances, particularly pooled CRISPR-based perturbation screens with high-content readouts and artificial intelligence/machine learning (AI/ML) models, are making this goal tractable. The text describes how these technologies enable researchers to systematically map genetic circuits, predict unseen cellular states and perturbation outcomes, and iteratively refine these predictions through active learning experiments, ultimately aiming to unify various levels of cell biology.
References:
