DNA begins falling apart the moment a cell dies, but how quickly it disappears depends on a long list of conditions. Under the harshest circumstances, exposed DNA can become unreadable in weeks; under the best, recoverable fragments have survived for more than a million years in permafrost sediments. A landmark study of 158 radiocarbon-dated moa bones put the average half-life of a short mitochondrial DNA sequence at roughly 521 years, meaning that after that span, about half the bonds holding a given stretch together will have broken. That number, though, is a single data point from one set of conditions in New Zealand, and real-world survival swings wildly depending on temperature, moisture, ultraviolet light, the biological tissue involved, and the minerals surrounding the molecule.
How DNA Falls Apart on Its Own
Even inside a living cell, DNA is not perfectly stable. The backbone of the molecule undergoes spontaneous hydrolysis, where water molecules attack the chemical bonds that link nucleotides together. Under conditions resembling those inside a cell, the half-life of the phosphodiester bond that forms the backbone is estimated at around 30 million years, so the backbone itself is remarkably tough. The real vulnerability lies elsewhere: in the bonds connecting the individual bases to the sugar-phosphate backbone and in chemical changes to the bases themselves.1PubMed Central. An Overview of Chemical Processes That Damage Cellular DNA: Spontaneous Hydrolysis, Alkylation, and Reactions with Radicals
The most damaging spontaneous reaction is depurination, in which the purine bases (adenine and guanine) pop off the backbone. This happens with a half-life of about 730 years per site under physiological conditions. The pyrimidine bases (cytosine and thymine) are lost more slowly, with a half-life closer to 14,700 years. Inside a living human cell, depurination alone generates roughly 10,000 abasic sites every day, but repair enzymes patch them almost as fast as they appear.1PubMed Central. An Overview of Chemical Processes That Damage Cellular DNA: Spontaneous Hydrolysis, Alkylation, and Reactions with Radicals Once the organism dies and those repair systems shut down, the damage accumulates unchecked.
Another chemical change, cytosine deamination, converts cytosine into uracil, which does not belong in DNA. This reaction proceeds with a half-life of roughly 30,000 to 85,000 years in double-stranded DNA at body temperature. It sounds slow, but over centuries it leaves a characteristic signature: ancient DNA fragments are littered with apparent C-to-T substitutions, especially at the ends of the molecules where the double helix has frayed into single strands.2PubMed Central. Ancient DNA damage Researchers studying Neandertal genomes have shown that these deamination-driven substitutions cluster dramatically at fragment tips, providing a kind of chemical fingerprint that distinguishes genuinely ancient sequences from modern contamination.3PubMed Central. Patterns of damage in genomic DNA sequences from a Neandertal
Temperature Is the Single Biggest Factor
If you had to pick one variable that predicts how long DNA will survive, it would be temperature. A large meta-analysis of ancient DNA samples confirmed that cytosine deamination is strongly correlated with both the age of a sample and the average temperature of the site where it was buried. Higher temperatures speed up the hydrolytic reactions that break bonds and scramble bases; cooler temperatures slow them down dramatically.4PubMed Central. A new model for ancient DNA decay based on paleogenomic meta-analysis This is why the oldest recoverable DNA sequences come from permafrost and high-latitude cave sediments, not from tropical sites. Paleogenomic work can now reach into the early Pleistocene, over a million years ago, but only in cold environments where the thermal clock effectively ran in slow motion.5Science. Deep-time paleogenomics and the limits of DNA survival
The 521-year half-life measured in New Zealand moa bones reflects a temperate climate with an average site temperature around 13°C.6PubMed Central. The half-life of DNA in bone: measuring decay kinetics in 158 dated fossils Move that same bone to a tropical environment and the rate of decay would accelerate. Move it to Siberian permafrost and the half-life would stretch considerably. The relationship between temperature and decay rate is not linear: it follows an exponential curve, so even a few degrees of difference compound into large effects over thousands of years.
Moisture, UV Light, and Microbial Attack
Temperature gets most of the attention, but humidity is a close second. Atmospheric water drives both strand breakage and oxidation of the bases. Studies on insect DNA show that degradation is worst when high humidity and high temperature combine: specimens stored at 90% relative humidity and 35°C lost DNA integrity far faster than those kept cool and dry.7PubMed Central. Effect of temperature and humidity on insect DNA integrity evaluated by real-time PCR Research on solid-state DNA degradation has confirmed that atmospheric water is a major driving force for strand scission and oxidation, even in dried samples stored at room temperature.8Materials Today Advances. DNA preservation under ambient conditions: degradation mechanisms, integrity assessment and storage strategies
Ultraviolet radiation presents yet another threat, though mainly to DNA at or near surfaces. UV-B light creates pyrimidine dimers, where two adjacent bases on the same strand fuse together, distorting the helix. It also causes direct strand breaks.9PubMed Central. Molecular mechanisms of ultraviolet radiation-induced DNA damage and repair In practice, UV damage matters most for forensic samples left exposed at crime scenes, for environmental DNA floating in shallow water, and for museum specimens displayed under lights. DNA buried underground or locked inside bone is largely shielded from UV.
Then there are living organisms. Right after death, a body’s own enzymes, including nucleases that specifically target DNA, begin chopping up the genome. This self-digestion, or autolysis, is the first and fastest wave of DNA destruction, outpacing the slower chemical processes of hydrolysis and oxidation by a wide margin.10ScienceDirect (Elsevier). The persistence of human DNA in soil following surface decomposition After the body’s own enzymes, soil bacteria and fungi continue the job. In warm, moist, biologically active soil, soft tissue DNA can become undetectable within months. In sterile, cold, or very dry environments, it lingers far longer.
Why Bone and Teeth Preserve DNA So Well
If you have ever wondered why ancient DNA studies so often involve teeth and dense bone, the answer lies in a mineral called hydroxyapatite. Bone and tooth enamel are composites of protein fibers and tiny hydroxyapatite crystals, and DNA has a specific chemical affinity for those crystals. When DNA adsorbs onto hydroxyapatite, it becomes more resistant to decay and less vulnerable to attack by nucleases and serum enzymes.11PubMed. DNA binding to hydroxyapatite: a potential mechanism for preservation of microbial DNA Molecular simulations confirm that the DNA double helix remains structurally stable when bound to hydroxyapatite surfaces or encapsulated inside nanoscale pores in the mineral.12PubMed. Biominerals Formed by DNA and Calcium Oxalate or Hydroxyapatite: A Comparative Study
The mineral matrix essentially wraps the DNA in a protective cage, slowing water access and blocking enzymes. Comparisons of 69 archaeological cattle bones and teeth found that the survival of ancient DNA in bone or dentine correlated with the survival of certain proteins, and that the ability of DNA to associate with hydroxyapatite crystals appeared to matter more than its association with proteins.13PubMed Central. Comparing ancient DNA survival and proteome content in 69 archaeological cattle tooth and bone samples from multiple European sites The petrous bone, a tiny pyramid-shaped mass of extremely dense bone in the skull’s temporal region, has become the gold standard for ancient DNA extraction because its thick mineral walls are among the best natural shields against degradation.
Mineralogy matters beyond bone, too. Environmental DNA shed by organisms into soil, lake sediments, and cave floors survives differently depending on what minerals it encounters. Surface charge, surface texture, and mineral composition all influence how strongly DNA adsorbs and how well it is preserved.14Environmental DNA. Survival of environmental DNA in sediments: Mineralogic control on DNA taphonomy Clay minerals, for instance, are well known to bind DNA tightly and slow its breakdown, which is why sediment cores from lake bottoms have become a major source of ancient environmental DNA.
The Amber Myth
If you grew up watching a certain dinosaur movie, you probably carry the idea that DNA trapped in amber can survive for tens of millions of years. The science does not support this. Multiple attempts to extract ancient DNA from insects preserved in amber that is 40,000 years old or older have failed or produced results that could not be replicated. Resin-embedded specimens are now generally regarded as unsuitable for genetic studies at deep timescales. Researchers have confirmed that DNA does persist in beetles embedded in six-year-old and two-year-old resin from Madagascar, proving that resin does not immediately destroy DNA, but that is a far cry from the millions of years amber would need to preserve it.15PubMed Central. DNA from resin-embedded organisms: Past, present and future
The problem is that resin, and the amber it eventually becomes, does not stop hydrolysis and oxidation the way extreme cold does. Over geological timescales, the same chemical reactions that degrade DNA everywhere else continue inside the amber, just possibly at a slightly reduced rate. The real preservative champions remain cold and dryness, not entombment in tree sap.
Mitochondrial DNA Versus Nuclear DNA
When researchers talk about extracting “ancient DNA,” they are not always talking about the same molecule. Each human cell has one copy of the nuclear genome but hundreds or even thousands of copies of the much smaller mitochondrial genome. That sheer copy-number advantage means mitochondrial DNA is far more likely to survive in degraded samples. In many archaeological and forensic cases where nuclear DNA is completely gone, mitochondrial sequences can still be recovered and analyzed.16PubMed Central. Mitochondrial DNA, a Powerful Tool to Decipher Ancient Human Civilization from Domestication to Music, and to Uncover Historical Murder Cases
This has practical consequences. Mitochondrial DNA can tell you about maternal lineage but not about the full genome, sex, or many traits governed by nuclear genes. For forensic identification of badly degraded remains, mitochondrial DNA often provides the last line of evidence when other methods have failed. For paleogenomic research, the push to recover nuclear DNA from ancient remains has required increasingly sophisticated laboratory techniques.
How Modern Technology Pushes the Limits
One of the reasons the field of ancient DNA has exploded in the past decade is that laboratory methods have gotten dramatically better at working with tiny, broken fragments. Traditional DNA library preparation methods, designed for high-quality modern DNA, lost a huge proportion of the ultra-short fragments that are all that remain in very old specimens. Single-stranded library preparation methods changed the game. By processing each strand of a DNA fragment independently, these protocols greatly improve the recovery of fragments shorter than 50 base pairs, which is exactly the size range that dominates ancient samples.17PubMed Central. Single-stranded DNA library preparation from highly degraded DNA using T4 DNA ligase The approach also increases the proportion of endogenous DNA relative to bacterial contamination, which is a constant problem with ancient specimens.18PubMed. Manual and automated preparation of single-stranded DNA libraries for the sequencing of DNA from ancient biological remains and other sources of highly degraded DNA
Contamination from modern human handlers has always been a headache in ancient DNA work. New computational tools now exploit the very damage patterns that make ancient DNA difficult to read. Since ancient fragments show predictable C-to-T substitutions at their ends from cytosine deamination, software can use those patterns to distinguish genuinely ancient sequences from modern contaminating DNA, estimating contamination rates and filtering out the noise.19PubMed Central. AuthentiCT: a model of ancient DNA damage to estimate the proportion of present-day DNA contamination The damage that destroys information also, paradoxically, authenticates what remains.
Forensic and Medical Implications of DNA Degradation
For forensic scientists, DNA degradation is not just an abstract problem but a daily practical challenge. As DNA breaks into shorter and shorter fragments, the maximum size of the sequence that can be copied and amplified shrinks. Standard forensic profiling methods that rely on amplifying relatively long stretches of DNA simply fail on badly degraded evidence.20PubMed Central. Analysis of Human Degraded DNA in Forensic Genetics This is why forensic labs have shifted toward mini-STR kits that target shorter fragments, and toward sequencing-based approaches that can work with what is available rather than demanding intact long stretches.
The timeline for forensic DNA recovery varies enormously depending on the scenario. Blood on a sidewalk exposed to sun and rain may yield no usable DNA after a few weeks. A bloodstain on clothing stored indoors at room temperature could remain typeable for years. Skeletal remains buried in temperate soil often yield results decades or even centuries after death, especially from teeth and the petrous bone. The same principles that govern ancient DNA survival apply: cooler, drier, more mineral-protected samples last longer.
Outside forensics, the behavior of degraded DNA has found medical applications. Cell-free DNA, the short fragments that circulate in blood after cells die naturally, is itself a product of the same degradation processes. Researchers have recognized that the fragmentation patterns of circulating cell-free DNA carry information about where the DNA came from, which has enabled advances in cancer diagnosis, organ transplant monitoring, and prenatal testing.21PubMed Central. Systematically Evaluating Cell‐Free DNA Fragmentation Patterns for Cancer Diagnosis and Enhanced Cancer Detection via Integrating Multiple Fragmentation Patterns In a sense, the same fragility that limits ancient DNA recovery has been turned into a diagnostic tool in living patients.
Why There Is No Single Expiration Date
People often want a clean answer: how many years does DNA last? The honest response is that the question is incomplete without specifying the conditions. A few rough benchmarks help frame the range. Under the worst conditions, exposed soft tissue in a hot, humid tropical environment with microbial activity and UV exposure, useful DNA can vanish within days to weeks. Under moderate indoor conditions, dried bloodstains or tissue samples stored at room temperature preserve amplifiable DNA for years to decades. In buried bone at temperate latitudes, DNA fragments typically survive for thousands to tens of thousands of years. In permafrost, the current record stretches beyond a million years.
The 521-year half-life from the moa study is sometimes misunderstood as a universal constant. It is not. It is specific to mitochondrial DNA in bone, from a particular site, at a particular temperature, and it describes the rate at which bonds break in a 242-base-pair stretch, not the point at which DNA becomes unreadable.6PubMed Central. The half-life of DNA in bone: measuring decay kinetics in 158 dated fossils A half-life means that after 521 years roughly half the bonds in that stretch are broken, but many fragments still survive. After several half-lives, the fragments get shorter and fewer, and eventually what remains is too degraded to sequence, but that final limit depends on the sensitivity of whatever technology is being used to read it. As sequencing methods improve, DNA that was previously considered “gone” becomes recoverable.
That moving target is part of what makes the field so dynamic. Twenty years ago, the practical limit for ancient DNA recovery was thought to be around 100,000 years. Ten years ago it was pushed to 700,000. Now sequences over a million years old have been reported from permafrost sediments. The molecule has not changed; the tools have. Whether future advances can push the boundary further depends on whether any intact fragments survive at all, or whether at some point chemistry wins completely and every last bond has broken. For samples stored in the coldest, driest, most mineral-rich conditions on Earth, that final expiration date, if it exists, has not yet been reached.