Interpretation, extrapolation and perturbation of single cells.

Clinical Snapshot

Key Findings

• Primary Outcome: Not applicable - methodological review

Clinical Application

Limited direct clinical application; primarily research methodology May inform research infrastructure and computational biology capabilities in Australian research institutions Researchers and computational biologists working with single-cell data

Abstract

Single-cell analyses have transitioned from descriptive atlasing towards inferring causal effects and mechanistic relationships that capture cellular logic. Technological advances and the growing scale of observational and interventional datasets have fuelled the development of machine learning methods aimed at identifying such dependencies and extrapolating perturbation effects. Here, we review and connect these approaches according to their modelling concepts (including representation learning, causal inference, mechanistic discovery, disentanglement and population tracing), underlying assumptions and downstream tasks. We propose a unifying ontology to guide practitioners in selecting the most suitable methods for a given biological question, with detailed technical descriptions provided in an online resource . Finally, we identify promising computational directions and underexplored data properties that could pave the way for future developments.