The Unstable Genome

Under a microscope, a human cell looks orderly. A nucleus, bounded and contained. Chromosomes, neatly partitioned. It gives the impression of control — of a system that knows exactly what it is doing.

It does not.

At any given moment, the genome inside that nucleus is being damaged thousands of times. Chemical bonds break. Bases are modified. Strands snap. Most of this is not the result of radiation or toxins, but of the cell’s own existence. Oxygen metabolism produces reactive species that attack DNA. Water, present in overwhelming abundance, slowly hydrolyses it. Replication machinery, precise but not infallible, makes mistakes.

Left alone, the genome would not last a day.

What allows life to persist is not stability, but repair.

A System Built on Surveillance

Cells do not passively tolerate damage. They actively search for it.

Proteins patrol the DNA, scanning for distortions in the double helix — subtle irregularities that signal something has gone wrong. A mismatched base pair. A missing nucleotide. A kink where there should be none. These are not large defects. They are changes on the scale of atoms, yet they are detected with remarkable fidelity.

This detection is not random. It is structured. Damage recognition proteins exploit the physical properties of DNA — its flexibility, its thermodynamic stability — to identify sites that deviate from the norm. Some slide along the helix, interrogating it continuously. Others bind transiently, sampling configurations until something appears incorrect.

Different classes of damage are handled by different systems. Small, chemically altered bases are removed and replaced through base excision repair. Larger distortions — those caused by ultraviolet light, for instance — are cut out in short stretches by nucleotide excision repair. Errors introduced during replication are corrected by mismatch repair, which identifies the newly synthesised strand and excises the incorrect base.

Double-strand breaks, the most dangerous lesions of all, are resolved by either stitching the ends back together directly or by using a homologous sequence as a template to reconstruct what was lost. The former is fast but error-prone. The latter is precise but constrained to certain phases of the cell cycle.

Repair, like everything else in biology, is conditional.

Replication Under Constraint

DNA replication is often described as accurate. It is more accurate than most chemical processes of comparable complexity, but it is not flawless.

Polymerases — the enzymes that copy DNA — incorporate nucleotides at high speed, adding thousands of bases per minute. They possess proofreading activity, able to detect and remove incorrectly paired nucleotides immediately after insertion. This reduces error rates substantially, but not completely.

What remains is handed off to mismatch repair systems, which operate after replication has occurred. They identify subtle asymmetries between the original and newly synthesised strands and correct errors that escaped proofreading.

Even with both layers in place, mutations persist. On the order of one error per billion bases per replication cycle. In a genome of three billion bases, that is not negligible.

Every cell division produces variation.

Why Imperfection Persists

If DNA repair were flawless, evolution would not exist.

Every time a cell divides, its genome is copied. Even with proofreading and repair, a small number of errors persist. Most are neutral. Some are harmful. A very small fraction confer an advantage in a particular environment. Over generations, these differences accumulate.

At the scale of populations, this accumulation becomes visible as adaptation. At the scale of molecules, it is simply the consequence of imperfect copying.

This creates a fundamental constraint. Repair systems cannot eliminate all error without also eliminating variation. And variation is the substrate on which natural selection acts.

The balance is not deliberate. It is the outcome of evolutionary pressures acting on systems that themselves arose through error. There is no optimal point in any absolute sense — only a dynamic equilibrium between competing demands.

Too much instability, and organisms fail to maintain integrity. Too little, and they fail to adapt.

Life exists in between.

Layers of Protection

The genome is not protected by repair alone.

Cells regulate when and how they divide, ensuring that damaged DNA is not propagated unchecked. Checkpoint pathways monitor the integrity of the genome and can halt the cell cycle if damage is detected. If repair is not possible, cells may enter senescence — a state of permanent arrest — or undergo programmed cell death.

These responses are not fail-safes in the engineering sense. They are additional layers of control, each with its own limitations.

Checkpoint activation depends on damage thresholds. Some lesions are tolerated. Others are missed entirely. Apoptosis removes damaged cells, but only if signalling pathways remain intact. Senescence prevents proliferation, but senescent cells accumulate and contribute to tissue dysfunction over time.

There is no single mechanism that guarantees genomic fidelity. There is only redundancy.

When Repair Becomes the Problem

The same systems that preserve the genome can, under certain conditions, contribute to its instability.

Error-prone repair pathways, such as non-homologous end joining, can introduce mutations while resolving double-strand breaks. In situations where rapid repair is prioritised over accuracy — for instance, in immune cells generating antibody diversity — this is not only tolerated but required.

In other contexts, it is detrimental.

Chronic activation of repair pathways can lead to genomic rearrangements. Repetitive sequences may be misaligned and incorrectly joined. Transposable elements — segments of DNA capable of moving within the genome — may become active under stress, inserting themselves into new locations and disrupting gene function.

The genome is not only repaired. It is reshaped.

When Repair Fails

Cancer is often described as uncontrolled growth. More precisely, it is a failure of genomic maintenance.

Mutations accumulate in genes that regulate the cell cycle, apoptosis, and DNA repair itself. A cell that can no longer accurately detect or correct damage begins to diverge genetically from its neighbours. It becomes, in effect, a population of one, subject to its own internal evolutionary pressures.

Selection acts within the tumour. Cells that proliferate faster, evade immune detection, or resist therapy outcompete others. The genome becomes increasingly unstable, accelerating the process.

This instability is not uniform. Different regions of the genome accumulate mutations at different rates. Structural variations — duplications, deletions, inversions — become common. Entire chromosomes may be gained or lost.

What emerges is not a single disease, but a dynamic system evolving in real time.

Ageing as Accumulation

Ageing does not result from a single type of damage, nor from a single failure in repair. It reflects accumulation.

Over time, repair systems decline in efficiency. Not abruptly, but gradually. Damage that would once have been corrected persists. Mutations accumulate. Epigenetic regulation — the system that controls gene expression without altering DNA sequence — becomes less precise.

Cells enter senescence more readily. Tissues lose regenerative capacity. Stem cell populations become depleted or dysfunctional.

The genome remains largely intact, but its fidelity erodes.

There is no clear boundary between maintenance and decline. The same processes that sustain life also contribute, over decades, to its deterioration.

A Dynamic Genome

What emerges from all of this is a different picture of the genome. Not as a fixed archive of information, but as something closer to a process.

It is written, copied, corrected, and occasionally miswritten again. Its integrity is not guaranteed, only maintained. The sequence that defines an organism is, at every moment, contingent — dependent on systems that are themselves subject to failure.

Even within a single individual, the genome is not uniform. Somatic mutations accumulate across tissues. Cells in the same organ may carry distinct genetic changes, invisible at the level of the whole organism but consequential at the level of individual cells.

Identity, at the molecular level, is not absolute. It is maintained.

The Cost of Fidelity

Maintaining the genome is energetically expensive. Repair requires enzymes, signalling pathways, checkpoints — all of which must be synthesised, regulated, and coordinated. The cell invests significant resources into preserving information that is, chemically speaking, inherently unstable.

This investment has limits.

Under conditions of stress — nutrient deprivation, oxidative damage, environmental insult — the balance shifts. Repair may be deprioritised. Errors increase. Systems that normally operate with high fidelity begin to tolerate approximation.

The genome reflects the conditions under which it is maintained.

What the Genome Actually Is

It is tempting to think of DNA as the defining feature of life — a stable code that persists across generations. But this is an abstraction.

In reality, DNA is a reactive molecule, constantly subject to chemical change. Its persistence is not a property of the molecule itself, but of the systems that act upon it.

The genome is not stable. It is stabilised.

And that distinction matters, because it reframes how we understand everything from evolution to disease. Life does not depend on perfect information. It depends on information that can be maintained, imperfectly, for long enough to matter.

What we inherit is not a pristine sequence, but a sequence that has survived — copied, corrected, and altered — across an unbroken chain of instability.

That it persists at all is not inevitable. It is the outcome of a system that has, for billions of years, managed to hold itself together just well enough.