The Lytic Cycle: A Cell-by-Cell Cellular Automaton

Five steps, endlessly repeated

The lytic cycle is the standard route a virus takes to reproduce: hijack a host cell's own machinery, force it to build thousands of viral copies, then destroy the cell to release them. It is usually described in five stages — attachment, penetration, replication, assembly and lysis — and while the molecular detail differs enormously between a bacteriophage infecting a bacterium and a virus like influenza infecting an animal cell, the abstract sequence is the same across nearly all lytic viruses.

1. ATTACHMENT   virion binds a specific receptor on the host cell surface
2. PENETRATION  the viral genome (and sometimes accessory proteins)
                enters the cell; the capsid may stay outside
3. REPLICATION  host machinery is redirected to copy the viral genome
                and translate viral proteins  -- the ECLIPSE PERIOD:
                no infectious virus is detectable inside the cell yet
4. ASSEMBLY     new genomes and capsid proteins self-assemble into
                complete virions
5. LYSIS        the cell membrane ruptures, releasing the BURST SIZE --
                often dozens to hundreds of new virions -- which go on
                to infect nearby cells

The eclipse period: present but invisible

Between penetration and the appearance of the first complete new virion lies the eclipse period — a span of time during which the virus is unambiguously active inside the cell, redirecting ribosomes and polymerases toward its own genome, yet no infectious viral particle can be recovered from the cell if you were to break it open and assay it. This is not the virus hiding; it is the virus existing in a disassembled, intermediate form — bare genome and separate protein components — that has not yet been packaged into an infectious unit. The eclipse period's length varies enormously by virus, from tens of minutes for some fast bacteriophages to many hours for more complex animal viruses, and it sets a hard lower bound on how quickly a single infected cell can start contributing new infectious particles to a population.

Burst size and the shape of a spreading infection

When lysis finally occurs, the number of virions released — the burst size — is the single most consequential parameter for how fast infection spreads through a population of cells. If a burst releases, on average, more than one virion capable of successfully infecting a new cell, the infection grows; if fewer than one succeeds on average, it dies out. This threshold is the cellular-scale analogue of the basic reproduction number, R0, used to describe epidemic spread at the population scale in models like SIR — the same branching-process logic, just operating cell-to-cell instead of person-to-person.

Modeling it as a cellular automaton

Because the process is spatial — an infected cell can only realistically infect cells physically near it, since diffusing virions have limited range and limited lifetime — a natural way to simulate it is a grid-based cellular automaton, where each cell occupies one of a small set of discrete states and progresses through them on a fixed or stochastic timer:

cell states: HEALTHY -> INFECTED (eclipse) -> PRODUCING -> LYSED -> (empty)

each tick:
  for every INFECTED/PRODUCING cell, advance its internal timer
  when the eclipse timer expires: cell becomes PRODUCING
  when the production timer expires: cell LYSES --
    scatter burst_size new virion particles into the neighbourhood,
    diffusing outward with some decay probability per step
  a virion particle reaching a HEALTHY cell infects it with some
    probability proportional to local virion concentration

This kind of spatial, discrete-state model captures something the classical well-mixed differential-equation models of virus dynamics (which assume every cell is equally exposed to every virion) miss entirely: local structure. An infection on a 2D sheet of tissue spreads as a roughly circular wavefront whose radius grows linearly in time rather than the exponential growth a well-mixed model would predict, because most virions released near the center of an already-infected region land on cells that are already infected or dead, and only virions near the expanding edge find fresh, healthy cells to infect. Spatial structure alone, independent of any change in the underlying biology, slows an epidemic down.

Why this matters beyond a single infection

The same eclipse-period-plus-burst-size structure underlies how virologists interpret plaque assays, the standard laboratory technique for counting infectious virus: a single virion infects one cell in a lawn of cultured cells, the lytic cycle repeats locally exactly as described above, and the visible clear "plaque" that results is literally the spatial wavefront pattern this simulation reproduces, grown large enough to see with the naked eye after a day or two of incubation.

Frequently asked questions

Try it live

Everything above runs in your browser — open Virus Lytic Cycle and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.