Evolvoid: a genetic algorithm for shaping optimal cellular constructs

Area of Science:

• Computational Biology
• Biophysics

Background:

• Accurate cellular models are crucial for tissue function and drug discovery.
• Current in vitro models (spheroids, organoids) are expensive, low-throughput, and lack reproducibility.
• Evolvoid is a computational pipeline using genetic algorithms to generate virtual 3D cell constructs.

Purpose of the Study:

• To develop a cost-effective and reproducible method for designing 3D cellular models.
• To optimize virtual cell construct morphologies by simulating biophysical principles.
• To reduce reliance on in vitro and animal models in research.

Main Methods:

• Integrates finite element simulations with evolutionary principles.
• Employs genetic algorithms to evolve populations of random shapes.
• Utilizes a fitness function based on surface energy minimization and cell viability under oxygen gradients.

Main Results:

• Evolvoid generates virtual 3D spheroid-like constructs with optimized morphologies.
• Fitness function incorporates thermodynamic principles, nutrient transport, and cell uptake.
• Evolved individuals exhibit shapes and sizes similar to in vitro spheroids.
• Shannon entropy tracks increasing genome complexity consistent with evolutionary dynamics.

Conclusions:

• Evolvoid is a modular, scalable, and tunable platform for in silico design of 3D cellular models.
• Enables the creation of high-fidelity digital twins of biological systems.
• Supports advancements in lab-on-a-laptop technologies and New Approach Methodologies (NAMs).