A Multimodal Workflow for Spatial Metabolic Neighborhood Mapping in Neural Rosette Cultures.

Area of Science:

• Biotechnology
• Stem Cell Biology
• Neuroscience

Background:

• Neural rosettes are key indicators of neural progenitor cells essential for central nervous system lineage development.
• Monitoring early differentiation changes, including positional and metabolic shifts, is crucial for optimizing stem cell culture and manufacturing.
• Current methods lack efficient ways to analyze the complex molecular and spatial data generated during differentiation.

Purpose of the Study:

• To develop an analytical framework for identifying neural rosettes from confocal microscopy images.
• To translate cell-resolved MALDI imaging data into interpretable readouts for cell manufacturing.
• To quantify molecular heterogeneity and spatial organization within differentiating stem cell colonies.

Main Methods:

• Utilized confocal microscopy to identify neural rosettes in differentiating stem cells.
• Applied a pipeline to convert regions of interest into single-cell feature matrices.
• Employed Principal Component Analysis (PCA), Leiden clustering, and Uniform Manifold Approximation and Projection (UMAP) for data summarization and visualization.
• Co-registered MALDI imaging data with microscopy for cell-resolved metabolic analysis.

Main Results:

• Developed a framework to analyze neural rosettes and associated metabolic shifts during stem cell differentiation.
• Generated metabolic neighborhoods that quantify molecular heterogeneity within cell colonies.
• Demonstrated that these metabolic neighborhoods form coherent spatial domains when mapped back to x-y space.
• Provided a method to compare metabolic neighborhoods across conditions and identify potential critical quality attributes.

Conclusions:

• The developed framework offers a practical bridge between multimodal MALDI imaging capabilities and process-relevant interpretation for cell manufacturing.
• Enables quantification of molecular heterogeneity and spatial organization without manual screening.
• Facilitates cell culture quality monitoring and optimization of batch culture yield for neural lineage development.