We seek to quantitatively describe the physical and chemical processes that define the functional state of a cell with computer modeling and simulation.

Our Goal

We are working to make 4D (space plus time) computer models of whole cells functioning under normal and stressed conditions. The modeling is guided by developing cutting-edge experimental tools that generate datasets with unprecedented resolution.

As to the computational component for the minimal bacterial cell, it ranges from Lattice Microbes simulations with 200,000 particles to the all-atom simulations with Martini/Gromacs for 2 billion particles. Only LM can simulate the entire cell cycle while the all-atom models are simulated around specific states of the cell, e.g., before, during, and after cell division.

First 4D Whole-Cell Model of Minimal Bacterium

Researchers at QCB, in collaboration with Johns Hopkins University, Harvard Medical School, and the J. Craig Venter Institute, achieved a landmark milestone in computational cell biology: the first complete 4D whole-cell simulation of an entire cell cycle in a living organism.​ This 4DWCM represents the most complete physics- and chemistry-based simulation of cellular life to date. It demonstrates that spatial heterogeneity is not merely a detail but a fundamental driver of cell cycle progression, gene expression variability, and division outcomes. The work was published in Cell.

Dive into the details

The Minimal Cell as a Model System

The study focuses on JCVI-syn3A, a genetically minimal synthetic bacterium with only 493 genes and a ~105-minute doubling time. Its stripped-down genome, abundant -omics data, and well-characterized structure make it the ideal testbed for a bottom-up, physics-based model of life. By simulating Syn3A from birth through division, the team aimed to uncover how spatial organization and molecular stochasticity govern cellular function.​

A Hybrid Computational Framework

The 4D Whole-Cell Model (4DWCM) integrates three simulation methods operating simultaneously:​

  • Reaction-Diffusion Master Equation (RDME) via Lattice Microbes — handles spatially resolved stochastic gene expression, ribosome diffusion, mRNA transcription and degradation, and membrane protein insertion
  • Ordinary Differential Equations (ODEs) — deterministically models the full metabolic network (glycolysis, nucleotide/lipid synthesis, transporters)
  • Brownian Dynamics (BD) via LAMMPS — simulates chromosome polymer dynamics, DNA replication, SMC loop extrusion, and topoisomerase activity

This hybrid approach spans processes ranging from nanometers to micrometers and from milliseconds to hours, all within a single unified cell-cycle simulation.​

Key Findings

  • Doubling time matched experiment: The model predicts a 105-minute cell cycle, in precise agreement with measured data, driven by lipid and membrane protein synthesis.​
  • DNA replication dynamics validated: Simulated origin-to-terminus (ori/ter) coverage ratios of 1.28 closely match the experimentally measured ratio of 1.21 from whole-genome sequencing.​
  • Stochastic cell-to-cell heterogeneity captured: Each of the 50 simulated replicate cells is unique — the model predicts not just average behavior but the full distribution of outcomes at division, including stochastic partitioning of ribosomes, degradosomes, and chromosomes to daughter cells.​
  • Gene expression sensitivity revealed: DNA replication initiation and the competition between mRNA translation and degradation are highly sensitive to spatial localization of molecular machinery — effects that well-stirred models cannot capture.​
  • New fluorescence imaging data: Airyscan confocal microscopy of 1,319 JCVI-syn3B cells expressing FtsZ-mCherry confirmed symmetric cell division morphologies, providing critical experimental constraints for the simulation.​

Significance

This 4DWCM represents the most complete physics- and chemistry-based simulation of cellular life to date. It demonstrates that spatial heterogeneity is not merely a detail but a fundamental driver of cell cycle progression, gene expression variability, and division outcomes. The framework and tools developed here — including Lattice Microbes, odeCELL, and btree_chromo — are being made publicly available and lay the foundation for next-generation whole-cell models that bridge molecular detail with cellular-scale behavior.​

 

 

 

Bringing Cells to Life via Minecraft

QCB’s CraftCells / Minecraft project is an innovative research-and-education effort that translates real 3D cellular imaging data and simulation outputs into interactive, voxel-based Minecraft worlds. By leveraging Minecraft’s native voxel environment, the team makes complex cell structures more accessible to students, researchers, and the public—allowing users to “enter” cells, explore organelle organization in 3D, and intuitively understand spatial relationships, scale, and morphology. The project features examples ranging from normal and cancer breast cells to neutrophils, yeast, and minimal-cell demonstrations, showing how advanced microscopy data can be transformed into immersive tools for scientific visualization, learning, and broader engagement. Learn more here.

Probing DNA, Protein, and Metabolite Interactiosn and Dynamics

Probing DNA, protein, and metabolite interactions and dynamics to uncover how molecular networks coordinate cellular function across space and time via MARTINI and GROMACS simulation.