Multi-scale modeling of GMP differentiation based on single-cell genealogies

Area of Science:

• Developmental Biology
• Systems Biology
• Computational Biology

Background:

• Hematopoiesis generates all blood cells from hematopoietic stem cells via differentiation.
• Existing models often use population-based approaches or single-cell resolution, but integrating them is key.
• Understanding differentiation mechanisms requires multi-scale modeling.

Purpose of the Study:

• To integrate population-based and single-cell models for inferring differentiation mechanisms.
• To study the differentiation of granulocyte-monocyte progenitors (GMPs) into granulocytes or monocytes.
• To investigate the time-dependent nature of lineage decisions in GMPs.

Main Methods:

• Developed a branching process model to infer differentiation probabilities from colony assays.
• Utilized single-cell time-lapse microscopy to determine differentiation genealogies.
• Established a stochastic toggle switch model with antagonistic transcription factors.
• Employed approximate Bayesian computing to infer model parameters.

Main Results:

• Differentiation probability in GMPs increases with cell generation, unlike predictions from standard branching models.
• A stochastic toggle switch model can explain both time-dependent and time-independent differentiation probabilities.
• Inferred parameters suggest distinct timescales for granulocyte and monocyte differentiation.
• Single-cell time-resolved observations are crucial for understanding cellular decisions.

Conclusions:

• Provides a multi-scale view of murine GMP differentiation.
• Highlights the inadequacy of simple branching models for capturing temporal dynamics.
• Emphasizes the necessity of single-cell, time-resolved data for mechanistic insights into hematopoiesis.