DNA is a long biopolymer built from repeating nucleotide monomers. Each nucleotide consists of a monosaccharide sugar called 2-deoxyribose, a phosphate group, and a nitrogenous nucleobase. The sugar lacks a hydroxyl group at the 2' carbon ring position, distinguishing it from the ribose found in RNA and conferring greater chemical resistance to hydrolysis.¹⁰¹¹ Adjacent nucleotides in a single strand are joined by covalent phosphodiester linkages between the 3' carbon of one sugar molecule and the 5' phosphate group of the next, forming an asymmetric, negatively charged sugar-phosphate backbone with a defined directional orientation from the 5' phosphate end to the 3' hydroxyl end.¹²

In living cells, DNA almost always exists as a double-stranded molecule in which two polynucleotide chains run in opposite, or antiparallel, directions. The hydrophilic sugar-phosphate backbones project outward toward the surrounding aqueous environment, while the hydrophobic nucleobases orient inward toward the central axis.¹³ The four nucleobases fall into two chemical classes: the double-ringed purines, comprising adenine and guanine, and the single-ringed pyrimidines, comprising thymine and cytosine. Hydrogen bonds form specifically between complementary bases on opposite strands, pairing adenine with thymine via two hydrogen bonds and guanine with cytosine via three hydrogen bonds.¹⁴ Because GC pairs possess an additional hydrogen bond and favorable aromatic base-stacking interactions, DNA regions with higher GC content exhibit greater thermodynamic stability and higher melting temperatures.

Under physiological conditions, the double helix predominantly adopts the right-handed B-DNA conformation.¹⁵ In this geometry, the helix measures approximately 2.0 nanometers (20 ångströms) in diameter and completes a full turn every 10.4 to 10.5 base pairs, spanning a helical pitch of 3.4 nanometers (34 ångströms). Because the glycosidic bonds attaching the base pairs to the sugar rings are not directly opposite one another, the twisting of the strands creates two spaces of unequal width along the exterior: a wider major groove and a narrower minor groove. Regulatory proteins, such as transcription factors, primarily bind within the major groove, where the exposed edges of the base pairs present distinct patterns of hydrogen-bond donors and acceptors that allow sequence-specific recognition without unwinding the helix.

DNA is a dynamic polymer capable of adopting alternative conformations depending on hydration, ion concentration, nucleotide sequence, and torsional strain. Under dehydrated conditions or within DNA-RNA hybrid duplexes, it forms A-DNA, a wider and flatter right-handed helix with tilted base pairs. Sequences rich in alternating purines and pyrimidines, particularly cytosine-guanine repeats under negative torsional strain, can transition into Z-DNA, a narrower left-handed helix with a zigzagging phosphate backbone that functions in transcriptional regulation and innate immune sensing. Additionally, DNA can form non-canonical architectures such as four-stranded G-quadruplexes in guanine-rich regions, cruciform structures at palindromic sequences, and four-way Holliday junctions during genetic recombination.

The total complement of DNA within a cell constitutes its genome. In prokaryotes, including bacteria and archaea, DNA is generally organized as a single circular chromosome located within the cytoplasm in a region called the nucleoid, often accompanied by smaller, independent circular DNA molecules known as plasmids.¹⁶ In eukaryotes, which include animals, plants, fungi, and protists, the vast majority of genomic DNA is arranged into multiple linear chromosomes housed inside a membrane-bound cell nucleus, while smaller, typically circular genomes reside within cytoplasmic organelles such as mitochondria and, in photosynthetic organisms, chloroplasts.

Because the total length of genomic DNA vastly exceeds the dimensions of the cell or nucleus, the molecule undergoes extensive topological compaction. The haploid human genome, for example, contains approximately 3.2 billion base pairs distributed across 23 chromosomes, or roughly 6.4 billion base pairs across 46 chromosomes in diploid somatic cells. In both prokaryotes and eukaryotes, enzymes called topoisomerases introduce or relax supercoils—over-winding (positive supercoiling) or under-winding (negative supercoiling) of the double helix—by transiently cleaving one or both DNA strands, thereby managing torsional stress and reducing the volume occupied by the molecule.

In eukaryotic nuclei, DNA is complexed with structural and regulatory proteins to form chromatin. The fundamental repeating unit of chromatin is the nucleosome, which consists of approximately 147 base pairs of DNA wrapped 1.65 times around a core octamer of alkaline proteins called histones, containing two copies each of histones H2A, H2B, H3, and H4. Strings of nucleosomes fold into higher-order chromatin fibers that are either loosely packed as transcriptionally accessible euchromatin or tightly condensed as transcriptionally silent heterochromatin, reaching maximum condensation as discrete chromosomes during cell division. The ends of linear eukaryotic chromosomes are capped by telomeres, specialized regions of repetitive hexanucleotide sequences (TTAGGG in humans) bound by the shelterin protein complex, which prevent the chromosome ends from being misidentified as double-strand breaks by cellular repair machinery.¹⁷

DNA performs its biological roles through the replication of its sequence prior to cell division and the regulated expression of its encoded genes. The framework governing information flow in cells, formulated by Francis Crick in 1958 as the central dogma of molecular biology, describes the directional transfer of sequential information from DNA to RNA and subsequently to protein.

The chemical isolation of DNA preceded the recognition of its biological function by three-quarters of a century. In 1869, working at the University of Tübingen, Swiss physician Friedrich Miescher isolated a phosphorus-rich, sulfur-free precipitate from the nuclei of human leukocytes collected from surgical bandages. Miescher named the substance "nuclein". Between 1878 and the 1890s, German biochemist Albrecht Kossel purified the non-protein component of nuclein—later termed nucleic acid—and identified its five primary nitrogenous bases (adenine, cytosine, guanine, thymine, and uracil) as well as chromatin histones.²⁵²⁶ In the early twentieth century, Lithuanian-American biochemist Phoebus Levene determined that nucleotides consist of a base, a sugar, and a phosphate linked in a specific order, and he distinguished between ribose and deoxyribose.²⁷ However, Levene also proposed the erroneous "tetranucleotide hypothesis," which held that DNA was a short, monotonous molecule composed of equal quantities of the four bases repeating in a fixed cycle, convincing most biologists that chemically diverse proteins, rather than nucleic acids, must be the carrier of heredity.

The case for DNA as the genetic material emerged from microbiology. In 1928, British bacteriologist Frederick Griffith observed that a heat-killed virulent strain of Streptococcus pneumoniae could convert a live, non-virulent strain into a lethal form, a phenomenon he called transformation. In 1944, Oswald Avery, Colin MacLeod, and Maclyn McCarty at the Rockefeller Institute systematically treated bacterial extracts with hydrolytic enzymes and demonstrated that DNA alone was the "transforming principle," as only deoxyribonuclease destroyed the transforming activity.²⁸ In 1952, Alfred Hershey and Martha Chase confirmed DNA's hereditary role using bacteriophage viruses labeled with radioactive phosphorus ($^{32}\text{P}$) for DNA and radioactive sulfur ($^{35}\text{S}$) for protein, proving that only viral DNA entered host Escherichia coli cells to direct viral reproduction. Concurrently, Austrian-American biochemist Erwin Chargaff used paper chromatography to refute the tetranucleotide hypothesis, establishing Chargaff's rules: while base composition varies across species, within any double-stranded DNA sample the molar amount of adenine equals that of thymine, and the amount of guanine equals that of cytosine.

Efforts to determine the three-dimensional structure of DNA centered on X-ray crystallography, pioneered on DNA fibers by William Astbury and Florence Bell in 1937.²⁹ In the early 1950s at King's College London, Rosalind Franklin, working with graduate student Raymond Gosling in the laboratory associated with Maurice Wilkins, produced high-resolution X-ray diffraction images of hydrated DNA fibers.³⁰ Their image of the hydrated B-form, known as Photograph 51, displayed a clear X-shaped diffraction pattern indicative of a helix.³¹ At the University of Cambridge, American geneticist James Watson and British biophysicist Francis Crick combined Franklin's crystallographic parameters—shown to them by Wilkins and through an institutional report—with Chargaff's base ratios to construct a physical model of the molecule. In April 1953, Watson and Crick published their two-chain, antiparallel, base-paired double-helix model in Nature, alongside supporting crystallographic papers by Wilkins and Franklin. Watson, Crick, and Wilkins shared the 1962 Nobel Prize in Physiology or Medicine; Franklin had died of ovarian cancer in 1958 at age 38 and was ineligible for the posthumous award.

Following the structural breakthrough, Arthur Kornberg isolated DNA polymerase in 1956, and Matthew Meselson and Franklin Stahl experimentally confirmed semiconservative replication in 1958. During the 1960s, Marshall Nirenberg, Har Gobind Khorana, Robert Holley, and Severo Ochoa deciphered the triplet codons of the genetic code.

Beginning in the 1970s, the discovery of bacterial restriction endonucleases and DNA ligases allowed scientists to cut and join DNA molecules from different species.³² In 1973, Stanley Cohen and Herbert Boyer developed recombinant DNA technology by inserting foreign DNA fragments into bacterial plasmids and cloning them in host bacteria, laying the foundation for the modern biotechnology industry.

In 1977, Frederick Sanger and Walter Gilbert independently devised methods to read the exact nucleotide order of DNA. Sanger's chain-termination method, utilizing dideoxynucleotides, became the standard for decades and was used to sequence the first complete DNA genome, that of bacteriophage $\Phi\text{X174}$. In 1983, Kary Mullis invented the polymerase chain reaction (PCR), a thermal cycling technique that uses heat-stable DNA polymerase and short synthetic primers to exponentially amplify a targeted DNA segment into millions of copies within hours.³³ Together, automated fluorescence sequencing and PCR enabled the international Human Genome Project, which launched in 1990, published a working draft alongside a private effort led by Celera Genomics in 2001, and released a reference human genome in 2003. Because complex repetitive regions near centromeres and telomeres remained unsequenced, the Telomere-to-Telomere (T2T) Consortium used long-read next-generation sequencing technologies to publish the first complete, gapless human genome sequence in March 2022.³⁴

In 2012, Jennifer Doudna and Emmanuelle Charpentier demonstrated that the bacterial adaptive immune system known as CRISPR-Cas9 (clustered regularly interspaced short palindromic repeats and CRISPR-associated protein 9) could be programmed with a synthetic guide RNA to cleave any targeted DNA sequence with high precision.³⁵ By harnessing cellular DNA repair pathways following cleavage, CRISPR-Cas9 allowed direct genome editing in living plant, animal, and human cells, earning Doudna and Charpentier the 2020 Nobel Prize in Chemistry. On December 8, 2023, the United States Food and Drug Administration (FDA), following earlier conditional authorization in the United Kingdom, approved exagamglogene autotemcel (Casgevy), the first clinical therapy utilizing CRISPR-Cas9 gene editing, to treat patients aged 12 and older with severe sickle cell disease by modifying the BCL11A enhancer in their hematopoietic stem cells.³⁶³⁷

Because human individuals—aside from monozygotic twins—possess unique variations in their DNA sequences, genetic analysis has become the gold standard for human identification since British geneticist Alec Jeffreys developed DNA fingerprinting in 1984.³⁸ Modern forensic DNA profiling relies on PCR amplification of short tandem repeats (STRs)—hypervariable non-coding regions where short nucleotide motifs repeat in varying numbers among individuals.³⁹ Law enforcement agencies compare crime scene STR profiles against national repositories such as the United States National DNA Index System, while organizations such as the Innocence Project have used post-conviction DNA testing to exonerate hundreds of wrongfully convicted prisoners.

In evolutionary biology, high-throughput sequencing of degraded ancient DNA (aDNA) recovered from subfossil bones and teeth has established the field of paleogenomics.⁴⁰⁴¹ Swedish geneticist Svante Pääbo pioneered rigorous contamination controls and single-stranded library preparation methods to sequence the first Neanderthal mitochondrial DNA in 1997 and a draft Neanderthal nuclear genome in 2010, before discovering a previously unknown archaic hominin group, the Denisovans, from a Siberian finger bone.⁴² Pääbo's work demonstrated extensive interbreeding between archaic hominins and ancestral modern humans and earned him the 2022 Nobel Prize in Physiology or Medicine. Similarly, the extraction of environmental DNA (eDNA)—traces of cellular genetic material shed by organisms into soil, water, or permafrost—allows ecologists to monitor contemporary biodiversity and paleontologists to reconstruct vanished ecosystems. In December 2022, researchers led by Eske Willerslev sequenced two-million-year-old eDNA preserved in frozen clay and quartz sediments from the Kap København Formation in northern Greenland, revealing an ancient boreal ecosystem inhabited by mastodons, reindeer, and poplar trees.⁴³

Beyond biology, the physical and informational properties of DNA are exploited in materials science and computing.⁴⁴ In DNA nanotechnology, researchers design synthetic oligonucleotides with complementary sequences that self-assemble through base pairing into programmable two- and three-dimensional nanoscale architectures, a technique known as DNA origami. In computer science, synthetic DNA is under development as an ultra-dense, low-energy medium for archival digital data storage. By converting binary code (0s and 1s) into nucleotide sequences (A, C, G, and T) synthesized onto short strands and read back via high-throughput sequencing, DNA storage offers a theoretical physical density exceeding 400 exabytes per cubic centimeter and durability spanning millennia when kept cold and dry, though high synthesis costs and slow read-write speeds remain active engineering challenges.

The proliferation of DNA technologies has generated persistent scientific, legal, and bioethical debates.⁴⁵ In forensic science, increased PCR sensitivity has enabled Low Copy Number (LCN) or low-template DNA profiling from microscopic traces of "touch DNA". Critics and defense experts have challenged the reliability of low-template profiles and complex multi-person DNA mixtures in court—such as in the English appellate cases R v Bates (2006) and R v Reed & Reed (2009)—citing stochastic amplification errors, allele drop-out ("voids"), secondary transfer, and divergent statistical results produced by competing probabilistic genotyping software programs.

Since the April 2018 arrest of the Golden State Killer, Joseph James DeAngelo, law enforcement agencies have increasingly used investigative genetic genealogy (IGG) to solve hundreds of cold cases.⁴⁶ By uploading crime scene single-nucleotide polymorphism (SNP) profiles to commercial or public genealogy databases, investigators identify distant relatives of a suspect and construct family trees. Civil liberties advocates and legal scholars have raised concerns that IGG subjects millions of people to genetic surveillance without their explicit consent, exposes sensitive health and non-paternity data, and lacks uniform statutory regulation across jurisdictions. Broader concerns regarding genetic discrimination by employers and insurers have prompted protective legislation in several countries, alongside resistance to proposals for mandatory universal DNA databases recorded at birth.

In medicine and agriculture, recombinant DNA and genome editing have sparked debates over biosafety, ecological impact, and equitable access. While somatic gene therapies such as Casgevy affect only the treated patient, they carry high financial costs, require intensive myeloablative chemotherapy, and necessitate long-term monitoring for off-target DNA cleavage or secondary malignancies.⁴⁷ Furthermore, the prospect of heritable germline genome editing in human embryos remains heavily restricted or prohibited worldwide due to safety risks and ethical objections to altering the human gene pool. Finally, within the history of science, the sociology of DNA research has drawn scrutiny regarding the historical under-recognition of Rosalind Franklin's experimental data during the 1950s and the later revocation of honorary titles from James Watson by Cold Spring Harbor Laboratory and the National Human Genome Research Institute following his scientifically unsupported public remarks regarding race and intelligence.