BOMA, a machine-learning framework for comparative gene expression analysis across brains and organoids.

Area of Science:

• Developmental Biology
• Computational Biology
• Genomics

Background:

• Understanding brain development is crucial for neuroscience and regenerative medicine.
• Organoids offer a model system to study human brain development, but their alignment with in vivo data requires robust methods.

Purpose of the Study:

• To develop and apply a machine learning framework, brain and organoid manifold alignment (BOMA), for comparing gene expression data between biological brains and organoids.
• To identify conserved and organoid-specific developmental trajectories and gene expression programs.

Main Methods:

• Global and local alignment of developmental gene expression data using manifold learning.
• Integration and analysis of single-cell RNA sequencing (scRNA-seq) data from human brains and organoids.
• Experimental validation of key gene expression findings using immunofluorescence.

Main Results:

• Human cortical organoids show gene expression alignment with specific cortical brain regions.
• Highly conserved gene expression patterns were identified between non-human primate and human brains around birth.
• BOMA revealed conserved and specific cell trajectories and clusters, with identified genes linked to brain- or organoid-specific functions.

Conclusions:

• BOMA effectively aligns developmental gene expression data, revealing conserved and specific biological processes.
• Human cortical organoids recapitulate region-specific gene expression programs.
• The framework provides insights into conserved developmental pathways and identifies novel targets for experimental validation.