In our idle game Cell Culture you unlock eight strains, each producing more biomass than the last. That progression is not arbitrary — it roughly follows how biologists actually climb the ladder of complexity, from a bacterium that divides every twenty minutes to a cell whose genome was written on a computer.
Here is what each of those eight really is, why it earns its place in the sequence, and one thing about it that is genuinely remarkable.
A rod-shaped bacterium, a normal resident of the mammalian gut, and probably the most thoroughly studied organism on Earth. It is the workhorse of molecular biology for one blunt reason: it is fast and it is cheap. Under favourable laboratory conditions a population can double roughly every twenty minutes, so an experiment you set up before lunch has results by the evening.
Its practical importance is hard to overstate. Insert a human gene into E. coli and the bacterium will dutifully manufacture the human protein. Recombinant human insulin — the first such drug approved for medical use, in the early 1980s — is made this way. Every time the game hands you cheap early biomass from E. coli, it is echoing why real labs start here too.
A single-celled fungus, and the first organism humans domesticated for chemistry rather than food: bread, beer, and wine all depend on it. Biologically, its significance is that it is a eukaryote — its DNA lives inside a nucleus, as ours does — while remaining as easy to grow as a bacterium.
That combination made it the bridge organism for understanding our own cells. Budding yeast was the first eukaryote to have its complete genome sequenced, in the mid-1990s, and work in yeast underpinned our understanding of how the cell division cycle is controlled — research recognised by the 2001 Nobel Prize in Physiology or Medicine. Many yeast genes are so similar to human ones that a human version can be swapped in and the yeast carries on living.
The first genuinely animal cell in the sequence. Fibroblasts are the maintenance crew of connective tissue: they secrete collagen and the rest of the extracellular matrix, the scaffolding that holds tissues in shape. When you cut yourself, fibroblasts migrate into the wound and lay down the matrix that becomes scar tissue.
They are also the classic starting material for cell reprogramming. A skin biopsy yields fibroblasts easily, they grow reliably in a dish, and — as the next entry explains — they can be persuaded to become something far more interesting.
A stem cell is defined by two abilities: it can renew itself indefinitely, and it can differentiate into other cell types. Pluripotent stem cells can, in principle, become any cell type in the body.
The landmark result here is induced pluripotency. In the mid-2000s, Shinya Yamanaka's group showed that introducing a small set of transcription factors into ordinary fibroblasts could reset them into a pluripotent state — turning a humble skin cell back into something embryo-like. It overturned the assumption that development is a one-way street, and shared the 2012 Nobel Prize with John Gurdon's earlier nuclear-transfer work. In the game, the jump in output at this tier reflects a real jump in capability: from a cell that does one job to a cell that can become anything.
Neurons trade in information. They maintain an electrical charge across their membrane and fire it as a travelling pulse — the action potential — passing signals to other neurons across synapses using chemical neurotransmitters. Current estimates put the number of neurons in an adult human brain at roughly 86 billion, with a far greater number of connections between them.
They are also awkward to culture, and that awkwardness is biologically meaningful: most mature neurons are post-mitotic — they have exited the cell cycle and no longer divide. A tissue that cannot simply replace its cells has to be repaired some other way, which is precisely why neurodegenerative disease is such a hard problem.
Cardiomyocytes contract, and they do it on their own schedule. Cardiac muscle is myogenic: the rhythm originates within the heart's own pacemaker cells rather than arriving as a command from the brain. Grow cardiomyocytes in a dish and a patch of them will spontaneously start beating — one of the more startling things to witness down a microscope.
Over a human lifetime the heart beats on the order of two to three billion times without a break. And like neurons, adult cardiomyocytes regenerate very poorly, which is why damage from a heart attack tends to be permanent — and why growing replacements from stem cells is such an active research goal.
Here the game stops counting cells and starts counting structures. An organoid is a miniature, simplified organ grown in three dimensions from stem cells. Given the right chemical cues and physical support, the cells do something remarkable: they self-organise, arranging themselves into the layered architecture of the tissue they came from, without being told where to go.
Intestinal organoids arrived around 2009, brain organoids a few years later, and the field has since produced miniature versions of liver, kidney, lung and more. Their value is that they behave far more like real tissue than a flat layer of cells in a dish — which makes them powerful tools for modelling disease and screening drugs, and a partial alternative to animal testing.
The final tier is the one that stops being biology-as-found and becomes biology-as-built. In 2010 the J. Craig Venter Institute assembled a complete bacterial genome from chemically synthesised DNA and transplanted it into a recipient cell, which then ran on the synthetic instructions. In 2016 the same group published a stripped-down version — a minimal cell carrying under 500 genes, close to the smallest genome known to support autonomous growth.
The most humbling detail: a substantial fraction of those essential genes have no known function. We can build a living cell from scratch and still not fully understand why every part of it is required. That gap is a fair summary of where synthetic biology stands.
Read the eight together and a pattern appears. The sequence moves from fast and simple to slow and sophisticated. Bacteria divide in minutes and need almost nothing; mammalian cells need controlled temperature, gas mixture and expensive media; neurons and cardiomyocytes barely divide at all; organoids need weeks of coaxing to assemble themselves. Biological capability is bought with time, cost and fragility.
That trade-off is, incidentally, what makes it work as an idle game. Early strains are cheap and let you build momentum; later ones cost enormously more but change the scale of what you can do. The economics of a petri dish turn out to map surprisingly well onto the economics of an incremental game.
Written for a general audience; details are simplified deliberately. Corrections are welcome via the contact page.