Boolean Network Models of Human Preimplantation Development.

Area of Science:

• Developmental Biology
• Computational Biology
• Genomics

Background:

• Single-cell transcriptomic studies are crucial for understanding human embryonic development and cell fate determination.
• Modeling these complex processes is challenging due to ethical and technical limitations of experimental perturbations.

Purpose of the Study:

• To develop an innovative computational method for modeling human embryonic differentiation using scRNAseq data.
• To generate stage-specific Boolean networks (BNs) by identifying pseudo-perturbations and integrating prior gene interaction knowledge.

Main Methods:

• Leveraging scRNAseq data from various human developmental stages.
• Identifying pseudo-perturbations in the absence of actual experimental perturbations.
• Integrating pseudo-perturbations with known gene interaction networks to construct Boolean networks (BNs).
• Applying the framework to medium and late trophectoderm developmental stages.

Main Results:

• Identified 20 pseudo-perturbations essential for inferring BNs.
• Generated BN families that delineate distinct regulatory mechanisms differentiating developmental stages.
• Demonstrated superior performance compared to existing pseudo-perturbation identification tools.

Conclusions:

• The developed framework effectively models human embryonic developmental processes and cell fate determination.
• The method provides insights into stage-specific regulatory mechanisms.
• This approach has potential applications across diverse developmental stages and research areas.