Reconstructing differentiation networks and their regulation from time series single-cell expression data.

Area of Science:

• Developmental Biology
• Computational Biology
• Genomics

Background:

• Accurate organogenesis modeling from single-cell RNA sequencing (scRNA-seq) data is challenging.
• Existing methods often assume gene expression similarity between parent and descendant cells, which is not always valid in vivo.
• In vivo studies involve complex, unsynchronized, and diverse cell populations, necessitating additional information for accurate lineage reconstruction.

Purpose of the Study:

• To develop a novel computational method for reconstructing dynamic developmental cell trajectories.
• To integrate gene expression similarity with regulatory information for improved organogenesis modeling.
• To identify key transcription factors (TFs) active during organogenesis and validate their role in cell fate determination.

Main Methods:

• Developed a probabilistic model that integrates gene expression similarity with regulatory information.
• Applied the method to single-cell RNA sequencing data from mouse lung development.
• Utilized existing and generated new experimental data for validation.

Main Results:

• The method accurately distinguished different cell types and lineages during mouse lung development.
• Successfully reconstructed dynamic developmental cell trajectories.
• Validated the identification of key regulators of cell fate.

Conclusions:

• The developed method enhances the accuracy of organogenesis modeling from scRNA-seq data by incorporating regulatory information.
• This approach overcomes limitations of methods relying solely on expression similarity.
• The findings provide a powerful tool for understanding developmental processes and identifying critical regulatory factors.