Back

Reconstitution of Human Brain Cell Diversity in Organoids via Four Protocols

Naas, J.; Balmana, M.; Holcik, L.; Novatchkova, M.; Dobnikar, L.; Krausgruber, T.; Ladstatter, S.; Bock, C.; von Haeseler, A.; Esk, C.; Knoblich, J. A.

2024-11-17 developmental biology
10.1101/2024.11.15.623576 bioRxiv
Show abstract

Human brain organoids are powerful in vitro models for brain development and disease. However, their variability can complicate use in biomedical research and drug discovery. Both the specific protocol as well as the pluripotent starting cell line influence organoid variability and can result in incomplete representation of brain cell types in an organoid experiment. Here, we systematically analyze the cellular and transcriptional landscape of brain organoids grown from multiple cell lines using four different protocols recapitulating dorsal and ventral forebrain, midbrain, and striatum. We establish the NEST-Score as a quantitative readout for cell line-driven and protocol-driven differentiation propensities by comparing cellular states across multiple cell lines and to in vivo reference data sets. Thereby, we establish a set of organoid protocols that together recreate the vast majority of cell types in the developing human brain and provide a reference for how well cell types are recapitulated across cell lines in each protocol. Additionally, we survey factors contributing to variability during organoid development and identify early gene expression signatures predicting protocol-driven organoid generation at later stages. We provide easy online access to our data through a web-based analysis tool, creating a reference for brain organoid research that allows rapid, straightforward validation of protocol and cell line performance.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.