The Statistical Analysis of Brain Organoid MEA Data

Advances in stem cell culture and bioengineering have enabled the generation of brain organoids from induced pluripotent stem cells. These three-dimensional (3D) structures, composed of thousands to millions of cells, mimic key features of human brain organization. Brain organoids are increasingly used to study neurodevelopment, model neurological disease (e.g., schizophrenia and bipolar disorder), and explore emerging biocomputing applications. A central focus of these studies is neurophysiological electrical activity, typically recorded over several minutes using multi-electrode arrays designed for in vitro systems. Statistical analysis of these data is challenging due to substantial noise and the need to characterize spatial and temporal structure of organoid activity. Statistical methods therefore play an essential role in separating signal from noise and in identifying temporal and spatial components and their interaction. In this manuscript, we review the analysis of brain organoid electrophysiology data, from raw data acquisition through current processing and analysis pipelines, and we highlight limitations of existing methods as well as opportunities for further statistical development. This dataset includes recordings from a 24-well MEA chip (MaxTwo).