Predicting pancreas cell fate decisions and reprogramming with a hierarchical multi-attractor model

Area of Science:

• Systems biology
• Cellular reprogramming
• Endocrinology

Background:

• Cell fate reprogramming, such as generating insulin-producing beta cells, relies on modulating transcription factors.
• Complex gene regulatory networks with feedback loops hinder prediction of reprogramming outcomes.
• Linear models are insufficient for predicting transcriptional perturbation effects.

Purpose of the Study:

• To develop a predictive model for cell fate reprogramming using qualitative gene regulatory interactions.
• To overcome limitations of incomplete regulatory network data for formal modeling.
• To guide the selection of transcription factors for targeted lineage reprogramming.

Main Methods:

• Developed a coarse-grained dynamical ordinary differential equation (ODE) based model.
• Utilized qualitatively described gene regulatory interactions as model input.
• Recapitulated existing cell state attractors (exocrine, beta, delta, alpha cells).

Main Results:

• The model successfully recapitulated observed cell type attractors.
• Predicted specific gene perturbations that can achieve desired lineage reprogramming.
• Demonstrated the predictive power of models based on incomplete network architectures.

Conclusions:

• Qualitative gene regulatory network information is sufficient for building predictive cell reprogramming models.
• The developed ODE model aids in making informed decisions for selecting transcription factors.
• This approach facilitates the design of effective cell fate reprogramming strategies.