Prediction of Growth Factor-Dependent Cleft Formation During Branching Morphogenesis Using A Dynamic Graph-Based...

Area of Science:

• Developmental Biology
• Computational Biology
• Bioengineering

Background:

• Branching morphogenesis is crucial for organ development, with epidermal growth factor (EGF) playing a key role in submandibular salivary gland (SMG) formation.
• Understanding cleft formation dynamics is essential for predicting gland morphology and function.
• Existing models often require extensive in vivo data or have high computational costs.

Purpose of the Study:

• To develop and validate a descriptive and predictive model for EGF-modulated branching morphogenesis in mouse SMG.
• To quantify the relationship between EGF concentration and cleft formation.
• To compare the model's predictive accuracy and computational efficiency against existing simulation methods.

Main Methods:

• Constructed a descriptive model using time-lapse videos to quantify ground truth via tissue-scale and local morphological features.
• Devised a predictive dynamic graph algorithm simulating EGF-driven branching morphogenesis using parameters like EGF concentration and mitosis rate.
• Compared simulation results with the Glazier-Graner-Hogeweg (GGH) model, assessing prediction accuracy and computational complexity.

Main Results:

• Higher EGF concentrations were correlated with increased bud formation and shallower cleft depths.
• The dynamic graph model accurately predicted SMG morphology, maintaining local structural characteristics.
• The model demonstrated comparable simulation accuracy to the GGH model but with significantly lower computational complexity.
• The enhanced model can predict SMG morphology for novel EGF concentrations without requiring ground truth time-lapse data.

Conclusions:

• The dynamic graph model provides an efficient and accurate in silico tool for studying branching morphogenesis.
• This approach reduces the need for extensive in vivo experiments, facilitating rapid testing of biological parameters.
• The model's ability to predict morphology without complete ground truth data represents a significant advancement.