Cell Type-Agnostic Transcriptomic Signatures Enable Uniform Comparisons of Neurodevelopment

Area of Science:

• Neuroscience
• Genomics
• Developmental Biology

Background:

• Single-cell transcriptomics advances understanding of neurodevelopmental cell types.
• Predicting cellular developmental stage from transcriptomic data remains difficult.

Purpose of the Study:

• To develop a robust method for predicting developmental age from cell transcriptomes.
• To establish a unified framework for comparing neurodevelopment across different contexts, model systems, and species.

Main Methods:

• Meta-analysis of over 2.8 million human developing brain cells.
• Training regularized regression models to infer cell-autonomous maturation.
• Developing a cell type-agnostic model for predicting developmental age.

Main Results:

• A cell type-agnostic model accurately predicts developmental age (2.6 weeks error) across diverse datasets and cell types.
• The model generalizes to human neural organoids, predicting normal and disease-induced developmental shifts.
• The model reveals an accelerated developmental tempo in the mouse brain compared to humans.

Conclusions:

• A unified, cell-agnostic approach accurately quantifies neurodevelopmental age from transcriptomic data.
• This framework enables robust comparisons of neurodevelopment across species and experimental models.
• The findings provide new insights into conserved and divergent patterns of brain development.