{
  "id": "dna",
  "title": "DNA",
  "lang": "en",
  "source_lang": "en",
  "summary": "DNA is the hereditary nucleic acid whose sequence stores biological information and whose replication, expression, and modification underpin cellular life.",
  "category": "science",
  "subcategory": "biology",
  "type": "concept",
  "aliases": [
    "Deoxyribonucleic acid",
    "Deoxyribose nucleic acid"
  ],
  "generated_at": "2026-10-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/dna/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/dna/",
      "json": "https://aiwiki.page/zh/dna.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "DNA, or deoxyribonucleic acid, is the nucleic acid that stores hereditary information in cellular organisms. It is a polymer assembled from nucleotides, usually forming two complementary strands arranged as a double helix. The sequence of its four principal bases—adenine, thymine, guanine, and cytosine—contains information used in cellular development, maintenance, and reproduction. DNA sequences include genes as well as regions involved in regulation and chromosome organization. (ncbi.nlm.nih.gov)\n\nComplementary base pairing enables DNA to be copied through DNA replication, while transcription makes RNA copies of selected sequences. DNA therefore connects the inheritance of a genome with the expression of biological functions. Its discovery and the elucidation of its structure transformed molecular biology; methods for reading and manipulating DNA subsequently made it central to genomic research and biotechnology. Understanding DNA encompasses its chemistry and physical organization, its interactions with cellular machinery, the processes that preserve or change its sequence, and the technologies used to investigate it. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Chemical composition and strand polarity",
      "text": "DNA is a nucleic acid whose strands are polymers of deoxyribonucleotide units. Each nucleotide contains a nitrogenous base, the five-carbon sugar 2-deoxyribose, and phosphate. The sugar’s carbon atoms are numbered 1′ through 5′; the prime marks distinguish sugar positions from the numbered atoms of the base. “2-Deoxy” identifies the replacement of the hydroxyl group at the 2′ position of ribose by hydrogen. A base joined to a sugar, without phosphate, is a nucleoside; adding phosphate produces a nucleotide. For example, deoxyadenosine is a nucleoside, whereas deoxyadenosine monophosphate, abbreviated dAMP, is a nucleotide. Free nucleotides can contain one, two, or three phosphate groups, although the repeating backbone linkages in DNA contain single phosphate groups. (ncbi.nlm.nih.gov)\n\nThe four principal DNA bases are adenine, guanine, cytosine, and thymine, abbreviated A, G, C, and T. Adenine and guanine are purines, with two fused rings; cytosine and thymine are pyrimidines, with one ring. The base is attached to the sugar’s 1′ carbon through a covalent β-N-glycosidic bond, involving nitrogen 9 of a purine or nitrogen 1 of a pyrimidine. Thus, the bases differ while the sugar–phosphate framework remains chemically repetitive. In sequence notation, A, G, C, and T identify the corresponding nucleotide residues within a strand, rather than isolated base molecules. The letters specify their order without depicting every atom or bond. (ncbi.nlm.nih.gov)\n\nAdjacent nucleotide units are joined by 3′–5′ phosphodiester bonds. Each bridging phosphate forms ester linkages to the oxygen at the 3′ position of one sugar and the oxygen at the 5′ position of the next. This arrangement creates an alternating sugar–phosphate backbone, with bases attached to the sugars rather than incorporated into the backbone itself. These covalent bonds maintain the continuity of an individual strand; they are distinct from the noncovalent interactions between paired strands. DNA synthesis uses the deoxyribonucleoside triphosphates dATP, dGTP, dCTP, and dTTP: incorporation retains the phosphate closest to the sugar in the backbone and releases the other two together as pyrophosphate. (ncbi.nlm.nih.gov)\n\nThe backbone gives a strand an intrinsic chemical direction, or polarity. At the 5′ end, the terminal sugar’s 5′ position is not linked to a preceding nucleotide; at the 3′ end, its 3′ position is not linked to a following nucleotide. A common end configuration is a 5′ phosphate and a free 3′ hydroxyl, although terminal chemical groups can vary without changing these directional labels. Sequences are conventionally written in the 5′→3′ direction: \\(5′\\text{-ACGT-}3′\\), for example, places the adenine-bearing residue at the 5′ side and the thymine-bearing residue at the 3′ side. Polarity describes backbone orientation, not a division into positively and negatively charged ends. (ncbi.nlm.nih.gov)\n\nRNA differs principally in containing ribose, with a 2′ hydroxyl group, and uracil in place of thymine among its four main bases. The 2′ hydroxyl provides a pathway for backbone cleavage unavailable to ordinary DNA: under alkaline conditions, its deprotonated oxygen can attack the adjacent phosphate, breaking the strand and forming a 2′,3′-cyclic phosphate intermediate. Subsequent hydrolysis yields 2′- or 3′-phosphate products. DNA’s lack of this hydroxyl makes its backbone more resistant to alkaline cleavage; this is a specific chemical distinction, not a claim that DNA cannot degrade. Strand number is likewise not an absolute discriminator: DNA is typically double-stranded and RNA often single-stranded, but single-stranded DNA and double-stranded RNA also exist. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "The double helix and complementary base pairing",
      "text": "In a DNA double helix, two polynucleotide strands wind around a common axis, with their sugar–phosphate backbones on the outside and their bases facing inward. The strands are antiparallel: when one runs in the 5′–3′ direction, its partner runs in the 3′–5′ direction. These labels refer to positions on the deoxyribose sugars, not to spatial directions such as “up” and “down.” In canonical Watson–Crick pairing, adenine pairs with thymine (A–T), and guanine pairs with cytosine (G–C). Each pair combines a two-ring purine with a one-ring pyrimidine, producing pairs of similar width that fit within a regular duplex. Their complementary shapes and arrangements of hydrogen-bond donors and acceptors account for this pairing specificity. (ncbi.nlm.nih.gov)\n\nAn A–T pair contains two hydrogen bonds, whereas a G–C pair contains three. These bonds join bases across the two strands; they are distinct from the covalent bonds connecting successive nucleotides within each backbone. Duplex stability also depends strongly on base stacking: neighboring, approximately planar bases overlap along the helix axis, allowing favorable noncovalent interactions. Consequently, counting hydrogen bonds alone does not explain the stability of a particular sequence; the identities and order of adjacent base pairs also matter. In 2006, Peter Yakovchuk, Ekaterina Protozanova, and Maxim Frank-Kamenetskii used DNA molecules containing individual nicks and gaps to separate stacking contributions experimentally, finding that stacking was the principal stabilizing contribution under the temperature and salt conditions they tested. (ncbi.nlm.nih.gov)\n\nB-DNA, the reference conformation for DNA under many hydrated conditions, is a right-handed helix approximately 2 nm in diameter. Successive base pairs are separated by an average axial rise of about 0.34 nm, with approximately 10.5 base pairs per complete turn. These values imply a mean rotation of roughly \\(360^\\circ/10.5 \\approx 34.3^\\circ\\) per base-pair step and a helical pitch of approximately \\(10.5 \\times 0.34 \\approx 3.6\\) nm. They describe average geometry rather than immutable dimensions: individual steps differ with sequence, and hydration, ionic conditions, and mechanical constraints can alter the structure. The base pairs lie approximately perpendicular to the helix axis, but local tilting and twisting make actual DNA less regular than an idealized molecular model. (ncbi.nlm.nih.gov)\n\nThe backbones are not equally spaced around the helix because the two sugar attachments to a base pair are not diametrically opposed. This asymmetry creates a wider major groove and a narrower minor groove, both spiraling along the duplex. The grooves expose the edges of the bases, enabling proteins to recognize sequence-dependent chemical features without first separating the strands. The major groove presents distinct patterns of hydrogen-bond donors, acceptors, and other groups for the four oriented pairs A–T, T–A, G–C, and C–G. The minor groove provides less chemically distinctive information, but its shape and electrostatic properties also support molecular recognition. In 2009, Remo Rohs and colleagues showed that narrow minor-groove regions frequently accommodate arginine residues in DNA-binding proteins, linking groove shape to sequence recognition. (ncbi.nlm.nih.gov)\n\nA base pair is one paired set of bases belonging to opposing strands and is also a unit of duplex length, abbreviated bp. A fully paired duplex of 1,000 bp contains 1,000 nucleotides in each strand—2,000 nucleotides altogether—and has an approximate B-DNA contour length of 340 nm. Contour length follows the molecular axis; it is not necessarily the distance between the ends of a bent molecule. Complementarity also determines sequence: opposite \\(5′\\text{–AGTC–}3′\\), the aligned partner is \\(3′\\text{–TCAG–}5′\\). Written in the conventional 5′–3′ direction, that partner becomes \\(5′\\text{–GACT–}3′\\), the reverse complement. Thus, specifying one strand’s sequence and polarity specifies its canonical partner, providing the template relationship used in DNA replication. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Alternative structures and DNA topology",
      "text": "B-DNA and A-DNA are right-handed double helices, but differ in base-pair inclination, sugar conformation, and dimensions. B-DNA predominates under many cellular conditions; A-DNA is broader and shorter, with approximately 11 base pairs per turn and an axial rise of 0.26 nm per pair. Reduced hydration favors A-DNA, although sequence and protein binding also influence the transition. These forms are families of conformations rather than rigid geometries. Z-DNA is left-handed, with a zigzag sugar–phosphate backbone and approximately 12 base pairs per turn. Alternating purine–pyrimidine sequences, particularly CG repeats, favor Z-DNA under suitable salt and torsional conditions. In 1979, Andrew H.-J. Wang, Alexander Rich, and colleagues established its atomic structure by X-ray crystallography of a CGCGCG fragment. (ncbi.nlm.nih.gov)\n\nOther conformations involve rearrangements of strand pairing. A hairpin forms when complementary regions within one strand pair, creating a double-stranded stem capped by an unpaired loop. In duplex DNA containing inverted repeats, both strands can fold into hairpins, producing a cruciform with a four-way junction. Negative supercoiling can supply the energy required for local strand opening and cruciform extrusion. In triplex DNA, a third strand associates with a duplex through additional base-specific hydrogen bonds in its major groove. Intramolecular H-DNA forms when part of a purine-rich or pyrimidine-rich mirror repeat folds back onto the remaining duplex; Sergei Mirkin and colleagues demonstrated the mirror-repeat requirement in 1987. (ncbi.nlm.nih.gov)\n\nA G-quadruplex consists of stacked guanine tetrads, each containing four guanines linked by Hoogsteen hydrogen bonding; potassium and other suitable ions stabilize its central channel. One strand can supply all four guanine-rich segments, or several strands can contribute them. Such structures are experimentally established, but a sequence capable of forming one does not necessarily remain folded in a cell. In 2013, Giulia Biffi and colleagues used the structure-specific BG4 antibody to visualize DNA G-quadruplexes in human cells. Proposed roles in gene regulation and replication require evidence at particular sites: structural detection alone does not establish a regulatory function, and probes or stabilizing ligands can alter the equilibrium being measured. (pmc.ncbi.nlm.nih.gov)\n\nDNA topology concerns constraints on strand interlinking and molecular deformation. Linear DNA with freely rotating ends can release torsional stress; covalently closed circular DNA cannot do so without strand breakage. Linear DNA can nevertheless behave as a topologically constrained domain when its ends or internal boundaries are rotationally restrained. For a closed duplex, the integer linking number, \\(Lk\\), counts the signed interlinking of its two strands:\n\n\\[Lk = Tw + Wr.\\]\n\nTwist, \\(Tw\\), describes winding of the strands around the duplex axis; writhe, \\(Wr\\), describes the axis’s spatial coiling. Neither must individually be an integer. Bending and twisting can redistribute them while preserving \\(Lk\\), provided the strands do not pass through one another. (pmc.ncbi.nlm.nih.gov)\n\nSupercoiling is defined relative to a torsionally relaxed reference state. For B-DNA containing \\(N\\) base pairs, \\(Lk_0 \\approx N/10.5\\); negative supercoiling has \\(\\Delta Lk=Lk-Lk_0<0\\), whereas positive supercoiling has \\(\\Delta Lk>0\\). For example, a 2,100-base-pair circle has an approximate relaxed reference of 200 turns; \\(Lk=190\\) gives \\(\\Delta Lk=-10\\). This deficit can be distributed between reduced twist and writhe, rather than appearing as exactly ten visible coils. Negative torsional stress can also favor local strand separation or transitions to cruciform, triplex, and Z-DNA structures. (ncbi.nlm.nih.gov)\n\nTopoisomerases change these constraints through transient, enzyme-controlled strand breaks followed by resealing. Type I enzymes cut one strand, permitting controlled rotation or single-strand passage, depending on the enzyme family. Type II enzymes cut both strands of one duplex segment, pass another duplex through the opening, and reseal it using an ATP-dependent cycle; this changes a closed duplex’s \\(Lk\\) in steps of two. Transient covalent protein–DNA bonds retain the cleaved ends during these reactions. Besides relieving torsional stress, type II enzymes separate interlinked DNA molecules—decatenation. In Escherichia coli, DNA gyrase introduces negative supercoils, whereas topoisomerase IV is principally responsible for decatenating replicated chromosomes. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Physical properties, strand separation, and hybridization",
      "text": "DNA is a negatively charged polymer: at approximately neutral pH, each internal phosphate group in its backbone carries roughly one negative elementary charge. Its behavior in solution therefore depends on surrounding ions, not simply on its sequence. Counterions such as \\(\\mathrm{Na}^{+}\\) and \\(\\mathrm{Mg}^{2+}\\) accumulate around DNA, while anions are partly excluded, creating an ion atmosphere that screens electrostatic repulsion. Increasing salt concentration within commonly studied aqueous conditions generally stabilizes a duplex by reducing repulsion between its two negatively charged backbones. Solvent composition also affects hydration, conformation, and solubility: adding ethanol in the presence of suitable counterions can promote aggregation and precipitation, whereas formamide destabilizes duplexes and permits strand separation at lower temperatures. Precipitation and denaturation are distinct processes; neither necessarily entails breaking DNA’s covalent backbone. (pubmed.ncbi.nlm.nih.gov)\n\nDuring thermal denaturation, increasing temperature disrupts the noncovalent interactions that maintain the duplex, including base pairing and stacking, allowing the strands to separate. The transition is often cooperative: opening one region facilitates opening neighboring regions, although long, heterogeneous molecules can melt in several sequence-dependent domains. The melting temperature, \\(T_{\\mathrm m}\\), is the midpoint of a specified melting transition. For a short duplex undergoing an approximately two-state transition, it corresponds to half the duplex population being dissociated; for long DNA, an optical midpoint need not mean that half the molecules are completely single-stranded. \\(T_{\\mathrm m}\\) is not a universal constant for DNA and must be reported with the relevant solution conditions. (pmc.ncbi.nlm.nih.gov)\n\nUnder otherwise comparable conditions, greater G–C content generally raises duplex stability, and longer short duplexes usually have higher melting temperatures than shorter ones. However, stability cannot be predicted merely by counting hydrogen bonds: adjacent base-pair identities, stacking interactions, terminal base pairs, and solvent conditions all contribute. John SantaLucia Jr.’s 1998 unified nearest-neighbor model represented sequence-dependent duplex thermodynamics through experimentally determined contributions from adjacent base-pair steps, together with initiation and salt corrections. For two separate complementary oligonucleotides, strand concentration also matters, because association depends on encounters between molecules. Mismatches generally destabilize a duplex, but their effects depend on the mismatched bases, neighboring sequence, and position; consequently, no single temperature penalty describes every mismatch. (pmc.ncbi.nlm.nih.gov)\n\nSeparated complementary strands can reassociate through renaturation, or annealing, when temperature and solution conditions favor duplex formation. Complementary regions first encounter one another and establish a short paired nucleus, after which pairing can extend along the strands. Julius Marmur and Paul Doty’s 1961 experiments demonstrated restoration of native-like DNA structure: for their preparations, optimal renaturation occurred approximately \\(25\\,^{\\circ}\\mathrm{C}\\) below \\(T_{\\mathrm m}\\), with extensive reassociation favored above \\(0.4\\,\\mathrm{M}\\) sodium ion concentration. These were experimental conditions, not universal requirements. Nucleic acid hybridization applies the same sequence-dependent association to strands from different sources, including DNA–DNA and DNA–RNA combinations. Higher temperature, lower salt, or more formamide can increase stringency, preferentially destabilizing imperfectly matched hybrids. Reannealing restores noncovalent structure; it does not repair chemically altered bases or reconnect a backbone broken by hydrolysis. (sciencedirect.com)\n\nDNA’s bases absorb ultraviolet light strongly near \\(260\\,\\mathrm{nm}\\). Stacking in the duplex suppresses this absorption relative to less ordered states; disruption of stacking during melting therefore increases absorbance, the hyperchromic effect. Recording \\(A_{260}\\) during heating produces a melting curve from which transition midpoints and changes in stability can be investigated. Absorbance also provides an approximate concentration measurement: for purified double-stranded DNA, \\(A_{260}=1\\) at a \\(1\\,\\mathrm{cm}\\) optical path conventionally corresponds to about \\(50\\,\\mu\\mathrm{g}/\\mathrm{mL}\\). This conversion depends on strandedness and sequence, and ultraviolet absorbance alone does not distinguish intact DNA from fragments, RNA, or other absorbing contaminants. (pmc.ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Distribution in cells, organelles, and viruses",
      "text": "In bacteria and archaea, chromosomal DNA occupies a nucleoid, a DNA-rich region of the cytoplasm without a surrounding nuclear membrane. The DNA is associated with proteins rather than existing as an unorganized, free molecule. In eukaryotes, most cellular DNA instead resides in the membrane-bounded nucleus, distributed among linear chromosomes. Many bacterial and archaeal chromosomes are circular, but circularity is not a defining property of prokaryotes. Escherichia coli has a circular chromosome, whereas Borrelia burgdorferi has a linear chromosome; linear chromosomes also occur in Streptomyces. Some bacteria have multiple chromosomes: Vibrio cholerae, for example, has two circular chromosomes. Thus, neither chromosome number nor molecular shape alone distinguishes prokaryotic from eukaryotic DNA organization. (ncbi.nlm.nih.gov)\n\nEukaryotic DNA also occurs outside the nucleus, notably in mitochondria and chloroplasts. Human mitochondrial DNA is a circular, double-stranded molecule whose standard reference sequence contains 16,569 base pairs. Its complete sequence was published in 1981 by S. Anderson and colleagues, including Frederick Sanger. It contains 37 genes: 13 encode protein components of oxidative phosphorylation, 22 encode transfer RNAs, and two encode ribosomal RNAs. These genes support mitochondrial energy metabolism and the organelle’s own protein-synthesis system, but they do not provide all the information needed to construct or maintain a mitochondrion. Many required proteins are encoded by nuclear DNA and imported into the organelle. (pubmed.ncbi.nlm.nih.gov)\n\nChloroplast genomes encode components of photosynthesis together with RNAs and proteins needed for gene expression within the organelle. Many land-plant chloroplast genomes are approximately 120–160 kilobases long and contain about 120 genes, although their sizes and gene complements vary. Circular genome maps do not imply that every physical DNA molecule is a simple circle: in 2004, Delene Oldenburg and Arnold Bendich reported predominantly linear and branched chloroplast DNA in maize seedlings. The retained genetic systems of mitochondria and chloroplasts, together with their relationships to bacterial genomes, support endosymbiotic ancestry. During organelle evolution, numerous genes were lost or transferred to nuclear DNA; consequently, organelle function depends on coordinated expression of nuclear and organellar genomes. (ncbi.nlm.nih.gov)\n\nPlasmids are DNA molecules maintained as replicating units distinct from the main chromosome. They are commonly circular and double-stranded, but linear plasmids also occur, including in Borrelia. Independent maintenance does not mean independence from the cell: plasmids use host machinery, and their replication depends on origins and associated control systems. Some can also integrate into a chromosome. Their accessory genes may confer advantages under particular conditions without being essential for ordinary growth. For example, the bacterial plasmid Rms312 carries determinants for resistance to tetracycline, chloramphenicol, streptomycin, sulfonamides, and mercury compounds. The E. coli F plasmid carries functions for conjugative DNA transfer, an example of horizontal gene transfer. (ncbi.nlm.nih.gov)\n\nDNA is the genomic material of many viruses, but not of every virus. Adenoviruses package linear, double-stranded DNA genomes of approximately 25–48 kilobases, whereas parvoviruses package linear, single-stranded DNA genomes of approximately 4–6 kilobases. Parvoviral single-stranded DNA is converted into double-stranded replicative forms inside host cells, illustrating the distinction between the genome packaged in a virus particle and nucleic-acid intermediates formed during infection. Other viruses have RNA genomes: influenza A virus, for example, packages eight segments of negative-sense, single-stranded RNA. Viral genomes therefore must be distinguished by nucleic-acid chemistry as well as strand number and molecular organization. (ictv.global)"
    },
    {
      "heading": "Chromosomes, chromatin, and genome packaging",
      "text": "A DNA molecule is a chemical polymer; a chromosome is an organized hereditary structure containing DNA together with associated proteins. In a typical unreplicated eukaryotic nuclear chromosome, one continuous double-stranded DNA molecule forms the molecular backbone. After replication, the chromosome comprises two sister chromatids, each containing its own DNA molecule and associated proteins, until their separation during cell division. The DNA–protein material is called chromatin: chromosomes remain chromosomes during interphase, when they are less compact and usually cannot be distinguished as individual rods by conventional microscopy. Chromatin packaging therefore describes both molecular composition and spatial organization, not merely the appearance of dividing chromosomes. (genome.gov)\n\nThe basic repeating unit of most eukaryotic chromatin is the nucleosome. Its core contains approximately 146–147 base pairs of DNA wrapped about 1.65 turns around a histone octamer containing two copies each of H2A, H2B, H3, and H4. Positively charged histone surfaces interact with the negatively charged DNA backbone, allowing extensive wrapping without breaking the DNA chain. Neighboring cores are connected by linker DNA of variable length; linker histone H1 associates near DNA entry and exit sites and promotes compaction. Nucleosome positioning and interactions influence which DNA regions are accessible to other proteins. Packaging is thus dynamic rather than a permanent enclosure of the sequence. (ncbi.nlm.nih.gov)\n\nNucleosomes do not invariably fold through a fixed hierarchy of regular fibers. The familiar 30-nanometer fiber can form under particular experimental conditions, but it is not a universal intermediate in living cells. In 2017, Horng Ou and colleagues used ChromEMT, combining DNA-selective staining with electron tomography, to examine human interphase nuclei and mitotic chromosomes. They observed irregular chromatin chains approximately 5–24 nanometers in diameter, with different particle arrangements and packing densities, rather than a uniform system of thicker fibers. These observations support descriptions of chromatin as a heterogeneous, flexible polymer whose local organization depends on nucleosome spacing, protein interactions, and cellular context. (pmc.ncbi.nlm.nih.gov)\n\nAt larger scales, interphase chromosomes form loops and contact domains—regions whose sequences interact more frequently with one another than with surrounding regions. The cohesin complex can generate loops by ATP-dependent extrusion: DNA is progressively drawn into an enlarging loop while the complex remains near its base. In 2019, Iain Davidson and colleagues directly demonstrated loop extrusion by purified human cohesin with NIPBL–MAU2, measuring rates up to 2.1 kilobase pairs per second under their assay conditions. During mitosis, condensin complexes help reorganize chromatin into compact, mechanically robust chromatids. Johan Gibcus and colleagues’ 2018 experiments in chicken cells supported condensin-dependent arrays of nested loops. Such experimentally supported mechanisms do not imply that every chromosome adopts an identical folding geometry throughout the cell cycle. (pubmed.ncbi.nlm.nih.gov)\n\nPackaging in bacteria and archaea differs from the canonical eukaryotic octamer-based system. Bacterial nucleoid-associated proteins can bend DNA and stabilize compact configurations: HU and IHF, for example, introduce substantial bends without assembling ordinary eukaryotic nucleosomes. Their effects depend on protein concentration and DNA structure. Many archaea instead possess histones homologous to eukaryotic core histones. Francesca Mattiroli and colleagues showed in 2017 that archaeal histone dimers can assemble into extended polymers around which DNA follows a nearly continuous superhelix. This variable-length organization contrasts with the discrete eight-histone nucleosome core; archaeal packaging systems are nevertheless diverse, and histones are not present in every archaeal lineage. (pubmed.ncbi.nlm.nih.gov)\n\nSpecialized chromatin also defines chromosome attachment sites and ends. A centromere supports assembly of the kinetochore, the protein structure connecting a chromosome to spindle microtubules for segregation. In humans, centromeric nucleosomes containing the H3 variant CENP-A help recruit this machinery. Telomeres protect the ends of linear chromosomes from inappropriate recognition and processing as broken DNA. Human telomeric DNA contains tandem TTAGGG repeats associated with the six-protein shelterin complex. Telomeric DNA can fold into a t-loop, in which the terminal region enters the duplex repeat tract; Jack Griffith and colleagues visualized such structures in human and mouse material in 1999. These protective structures are distinct from the mechanisms that replicate chromosome ends. (cmgm-new.stanford.edu)"
    },
    {
      "heading": "Sequence organization and the meaning of genetic information",
      "text": "A DNA sequence is the ordered succession of bases along a strand, conventionally written from 5′ to 3′; it may describe a short fragment, a gene, or an entire chromosome. A genome comprises the complete genetic material, including genes and sequences outside them. A locus is a specified genomic position or interval, whereas an allele is one of the alternative sequence versions found at that location. Thus, a locus identifies where a sequence occurs, not which version is present. Genes specify functional products: protein-coding genes encode RNAs that can be translated into proteins, while other genes produce functional RNAs, including transfer RNA and ribosomal RNA. Genetic information therefore includes both product sequences and sequences that influence their production. (genome.gov)\n\nA schematic, intron-containing eukaryotic protein-coding gene can be mapped in its direction of transcription as: promoter → transcription start → 5′ untranslated region → protein-coding region interrupted by introns → 3′ untranslated region → transcript-processing and termination region. The coding region begins at a translation-initiation codon and ends at a stop codon; its separated portions are joined in the mature messenger RNA. Exons are sequences retained in that RNA, whereas introns are removed by splicing. “Exon” is consequently not synonymous with “protein-coding”: untranslated regions, or UTRs, occupy exonic sequence but do not specify the main protein’s amino-acid sequence. Introns can also interrupt UTRs. UTR sequences can influence translation efficiency, RNA localization, and RNA stability. This schematic describes one common arrangement, not a universal gene architecture. (genome.gov)\n\nThe promoter provides a site at which transcription machinery assembles near the transcription start. Enhancers are regulatory sequences that bind proteins and can increase transcription from a target promoter; DNA folding can bring an enhancer and promoter into proximity despite their separation along the sequence. Regulatory elements may occur outside the transcribed region or within introns. Intergenic DNA lies between annotated genes, but “intergenic” does not mean “unregulated” or “without function.” Conversely, a gene’s regulatory sequences need not form a single continuous block immediately beside its coding region. (genome.gov)\n\nRepeated sequences have several distinct organizations. Tandem repeats place copies next to one another, as in CACACACA or larger satellite arrays; dispersed repeats occur at separated genomic locations. Many dispersed repeats descend from transposable elements, including LINEs, SINEs such as Alu, long-terminal-repeat retrotransposons, and DNA transposons. An identifiable element-derived sequence need not retain the ability to move. Genome composition differs markedly between organisms: Frederick Blattner and colleagues’ 1997 sequence of Escherichia coli K-12 contained 4,639,221 base pairs, with protein-coding genes occupying 87.8%. In contrast, only about 1.5–2% of human genomic DNA specifies protein sequence. The International Human Genome Sequencing Consortium’s 2001 draft analysis identified approximately 45% as transposon-derived repeats, a historical estimate dependent on the assembly and the ability to recognize diverged copies. (nature.com)\n\nNon-coding describes the absence of protein-coding capacity, not the absence of biological function. Functional RNA genes, promoters, enhancers, and untranslated regulatory sequences are all non-coding. Evidence of function must nevertheless be distinguished from simple biochemical activity. Detecting transcription, protein binding, or a chromatin-associated mark establishes an observable molecular event, but does not by itself demonstrate that the sequence is necessary for regulation or organismal fitness. Perturbation experiments can test whether changing or deleting a sequence alters expression or another phenotype in a specified context. Evolutionary conservation provides a different kind of evidence: unusually slow sequence change can indicate constraint imposed by natural selection. Lack of deep conservation does not exclude a recently evolved or lineage-specific function. A 2014 analysis by Manolis Kellis and colleagues explicitly distinguished biochemical, genetic, and evolutionary approaches, emphasizing that their different criteria need not identify identical sets of functional DNA. (genome.gov)"
    },
    {
      "heading": "DNA replication and chromosome-end maintenance",
      "text": "DNA replication is semiconservative: each daughter duplex contains one parental strand and one newly synthesized strand. Complementary base pairing makes each parental strand a template for assembling its partner. Replication begins at origins, where local DNA unwinding establishes Y-shaped replication forks, usually moving in opposite directions. DNA polymerases extend a pre-existing primer by adding deoxyribonucleoside triphosphates to its free 3′ hydroxyl group, releasing pyrophosphate as each phosphodiester bond forms. Consequently, synthesis proceeds only in the 5′–3′ direction, while the template is read 3′–5′. These enzymes cannot ordinarily initiate a strand without a primer. (ncbi.nlm.nih.gov)\n\nThe fork machinery coordinates unwinding, priming, and elongation. Helicases use nucleotide hydrolysis to separate the parental strands; single-strand-binding proteins—bacterial SSB or eukaryotic replication protein A, RPA—stabilize exposed DNA and limit secondary-structure formation. Primase synthesizes short RNA primers. Ring-shaped sliding clamps, exemplified by the bacterial β clamp and eukaryotic PCNA, retain polymerases on DNA and increase processivity; clamp loaders install these rings at primer–template junctions. In Escherichia coli, polymerase III performs most chromosomal synthesis. In eukaryotic nuclear replication, the polymerase α–primase complex produces an RNA primer followed by a short DNA extension; polymerase ε normally performs most leading-strand synthesis, whereas polymerase δ performs most lagging-strand synthesis. These proteins operate as an interacting replisome rather than as independent enzymes. (ncbi.nlm.nih.gov)\n\nBecause the templates are antiparallel, the leading strand is synthesized largely continuously toward the advancing fork, whereas the lagging strand is assembled from repeatedly primed Okazaki fragments extending away from it. Typical fragments measure approximately 1,000–2,000 nucleotides in bacteria and 100–200 in eukaryotes. In E. coli, polymerase I removes RNA primers through its 5′–3′ exonuclease activity and replaces them with DNA. Eukaryotic fragment maturation involves polymerase δ displacing primer-containing material into flaps that are cleaved by nucleases, notably FEN1. DNA ligase then seals the remaining nicks between adjacent fragments. Many replicative polymerases also perform proofreading: a mismatched newly incorporated nucleotide is transferred to a 3′–5′ exonuclease site, removed, and synthesis resumed. This differs from subsequent mismatch repair, discussed with other repair pathways. (ncbi.nlm.nih.gov)\n\nReplication organization differs between organisms. The circular E. coli chromosome normally initiates at one origin, oriC, whereas eukaryotic nuclear chromosomes use numerous origins during S phase of the cell cycle. Eukaryotic origin licensing loads inactive MCM helicase complexes before S phase; their later activation, coupled with inhibition of new licensing, prevents repeated copying of the same region within one cycle. Replication generally terminates when neighboring forks converge, after which synthesis is completed, nicks are sealed, and replication machinery is dismantled. Topoisomerases remove DNA interlinking that could obstruct daughter-chromosome separation. Experiments with Xenopus egg extracts published in 2019 additionally showed that topoisomerase II relieves topological stress during fork convergence itself. (ncbi.nlm.nih.gov)\n\nLinear chromosomes present an end-replication problem: after removal of a terminal lagging-strand primer, conventional polymerases lack an upstream 3′ hydroxyl from which to fill the gap. Processing newly replicated ends also contributes to terminal sequence loss. Telomerase counteracts this loss by extending the G-rich 3′ end of telomeric DNA. Its reverse-transcriptase subunit, TERT, copies a template within the enzyme’s own RNA component; repeated template realignment permits addition of successive repeats, including TTAGGG in humans. Conventional primer-dependent synthesis subsequently fills in much of the complementary strand. Carol Greider and Elizabeth Blackburn reported telomere-repeat-adding activity in Tetrahymena extracts in 1985. Telomere extension and end protection are distinct functions: mammalian shelterin proteins and telomeric structures prevent natural chromosome termini from eliciting inappropriate DNA-damage signaling or end-to-end fusion. Telomerase supplies repeat DNA; it does not by itself provide this protective cap. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Transcription and the expression of DNA information",
      "text": "During transcription, RNA polymerase locally separates the DNA strands and reads one—the template strand—in the \\(3′\\text{–}5′\\) direction, synthesizing complementary RNA in the \\(5′\\text{–}3′\\) direction. It adds ribonucleotides to the RNA’s 3′ end using ribonucleoside triphosphates, releasing pyrophosphate; unlike replicative DNA polymerases, it can initiate without a primer. The opposite, coding strand has the same sequence as the RNA, except that DNA contains thymine where RNA contains uracil. For example, a coding sequence \\(5′\\text{-ATGC-}3′\\), paired with template \\(3′\\text{-TACG-}5′\\), produces \\(5′\\text{-AUGC-}3′\\) RNA. Either DNA strand can serve as the template for different genes: promoter orientation determines which strand is read. DNA re-forms its double helix behind the advancing enzyme rather than being consumed. (ncbi.nlm.nih.gov)\n\nIn bacteria, initiation commonly involves a sigma factor that directs the core polymerase to a promoter. In Escherichia coli, promoters recognized by the housekeeping sigma factor typically contain sequence elements approximately 10 and 35 nucleotides upstream of the transcription start site. Polymerase opens the DNA, makes the first RNA bonds, and escapes the promoter into elongation. Termination releases the transcript and polymerase. Intrinsic termination commonly involves an RNA hairpin followed by a uridine-rich tract, which destabilizes the elongation complex; Rho-dependent termination uses the ATP-powered Rho protein to move along the nascent RNA and disrupt a paused transcription complex. These mechanisms end RNA synthesis, not protein synthesis: a transcription terminator is distinct from a translation stop codon. (ncbi.nlm.nih.gov)\n\nIn the nucleus of eukaryotic cells, RNA polymerase II transcribes protein-coding genes, whereas polymerases I and III produce major classes of ribosomal and transfer RNAs. Polymerase II initiation requires general transcription factors assembled at the promoter; TFIIH helps open DNA and phosphorylates the polymerase’s carboxy-terminal domain during the transition toward elongation. (ncbi.nlm.nih.gov) For most protein-coding transcripts, termination is coupled to RNA cleavage near a polyadenylation signal. Polymerase continues beyond the cleavage site; the downstream RNA is then degraded by a \\(5′\\text{–}3′\\) exonuclease, XRN2 in humans, helping terminate transcription. Steven West, Natalia Gromak, and Nick Proudfoot demonstrated XRN2’s involvement in human transcription termination in 2004. Thus, the mature RNA’s endpoint need not coincide with where polymerase leaves DNA. (nature.com)\n\nThe initial polymerase II transcript generally undergoes 5′ capping, splicing to remove introns and join exons, and addition of a 3′ poly(A) tail. Much of this processing occurs while transcription is still underway; mature messenger RNA is subsequently exported for translation. These steps process the RNA copy rather than rewrite its DNA template. (ncbi.nlm.nih.gov) During translation, a ribosome reads mRNA in three-nucleotide codons, while transfer RNAs supply corresponding amino acids. The standard genetic code has 64 codons: 61 specify amino acids and three specify termination. DNA therefore supplies the sequence template indirectly; it does not itself enter the ribosome as the message. Transcripts such as ribosomal and transfer RNAs function as RNA rather than being translated into proteins. (ncbi.nlm.nih.gov)\n\nThe central dogma concerns transfer of sequence information, not an obligatory DNA-to-RNA-to-protein itinerary for every gene. Francis Crick proposed it in 1958 and clarified it in 1970: information can pass between nucleic acids, but amino-acid sequence is not transferred back into nucleic-acid sequence. (dna.caltech.edu) Reverse transcription is therefore compatible with this formulation. In 1970, David Baltimore and, independently, Howard Temin and Satoshi Mizutani reported viral enzymes that synthesize DNA from RNA templates. (nature.com) Proteins can regulate transcription or catalyze DNA synthesis without serving as sequence templates. The restriction is not a prohibition on such chemical influences, but the absence of an established mechanism that reads a protein’s amino-acid order to reconstruct a corresponding nucleotide sequence. (dna.caltech.edu)"
    },
    {
      "heading": "DNA recognition, methylation, and epigenetic regulation",
      "text": "Sequence-specific molecular recognition allows transcription factors and restriction enzymes to distinguish particular DNA sites. In direct base readout, amino-acid side chains contact exposed chemical groups on bases, often through hydrogen bonds in the major groove. Shape readout instead detects sequence-dependent groove width, bending, deformability, and electrostatic potential; the bases responsible need not contact the protein directly. These mechanisms can operate together. In 2009, Remo Rohs and colleagues showed that narrow minor grooves concentrate negative electrostatic potential and preferentially accommodate positively charged arginine residues. In their structural dataset, arginine accounted for 60% of amino-acid contacts in narrow minor grooves, compared with 22% in wider grooves. Thus, recognition depends on the three-dimensional properties generated by a sequence, not simply its letters. Restriction-enzyme recognition can also be affected by chemical modifications within the binding site. (rohslab.usc.edu)\n\nDNA methylation adds methyl groups to bases without replacing the underlying nucleotide sequence. In bacterial restriction–modification systems, a methyltransferase marks host recognition sites, protecting them from the corresponding restriction enzyme, while unprotected incoming DNA may be cleaved. For example, the Escherichia coli EcoKI system attacks DNA lacking the appropriate adenine methylation. Other bacterial methyltransferases regulate DNA–protein interactions: E. coli Dam methylates adenine at its \\(N^6\\) position in 5′-GATC-3′ sequences. At the pap operon, which encodes pili, alternative methylation patterns interact with the regulatory proteins Lrp and PapI to produce reversible transcriptional ON and OFF states. Experiments by Aaron Hernday, Bruce Braaten, and David Low in 2003 showed that methylation of a promoter-proximal GATC site blocks PapI-dependent Lrp binding there, facilitating the ON state. (neb.com)\n\nIn mammals, cytosine methylation predominantly produces 5-methylcytosine at CpG sites, where cytosine is followed by guanine on the same strand. Its regulatory effects depend on genomic position and the proteins present, rather than constituting a universal “off” signal. Methylation can prevent particular transcription factors from binding or recruit methylated-DNA-binding proteins and associated repressors. In 1998, Xinsheng Nan and colleagues demonstrated that MeCP2-mediated repression involves a complex containing histone deacetylases, connecting DNA modification to chromatin regulation. Modifications of histones likewise change access to DNA without changing its sequence. Acetylation of histone lysines can alter histone interactions and increase local accessibility; methylation creates residue-specific binding surfaces. For example, methylation of histone H3 lysine 9 recruits HP1 proteins, which participate in transcriptionally repressive chromatin. (nature.com)\n\nCell-type-specific regulatory states combine transcription-factor binding, DNA methylation, and chromatin modifications. Ryan Lister and colleagues’ 2009 single-base-resolution maps identified extensive methylation differences between human embryonic stem cells and fetal fibroblasts. Some patterns persist through cell division by active maintenance rather than remaining chemically unchanged. After DNA replication, a previously methylated CpG can be hemimethylated: the parental strand carries the modification, whereas the newly synthesized strand does not. UHRF1 preferentially recognizes such sites and helps recruit the maintenance methyltransferase DNMT1 to restore methylation on the new strand. Magnolia Bostick and colleagues demonstrated UHRF1’s requirement for methylation maintenance in mammalian cells in 2007. This mechanism supports mitotic epigenetic memory, but does not imply that every modification is stably inherited. (pmc.ncbi.nlm.nih.gov)\n\nGenomic imprinting produces parent-of-origin-dependent expression of particular alleles. At the mammalian H19–IGF2 region, the unmethylated maternal imprinting control region binds CTCF, creating an enhancer-blocking boundary that restricts IGF2 activation. Paternal methylation prevents CTCF binding, allowing enhancer access to IGF2. Adam Bell and Gary Felsenfeld, and independently Amy Hark and colleagues, established this methylation-sensitive boundary mechanism in 2000. Persistence within somatic cell lineages must be distinguished from transmission across generations: mammalian germ-cell development and early embryogenesis involve extensive methylation reprogramming. Human embryo methylome studies published in 2014 documented broad demethylation after fertilization, with selective retention at some regions. Consequently, maintenance through mitosis alone is not evidence that an acquired regulatory state will pass through gametes to descendants. (nature.com)"
    },
    {
      "heading": "DNA damage, repair, and genome surveillance",
      "text": "A DNA lesion is a chemical or structural abnormality, whereas a mutation is a stable change in nucleotide sequence. A lesion may be repaired before it produces a mutation, block replication, or cause an incorrect sequence to become established during copying. Depurination hydrolyzes the bond attaching adenine or guanine to its sugar, leaving an abasic site; cytosine deamination produces uracil. Classic estimates place these spontaneous events at approximately 5,000 purine losses and 100 cytosine deaminations per human cell per day, rather than universal rates for every cell type. Reactive oxygen species modify bases and sugars, including formation of 8-oxoguanine. Ultraviolet radiation produces cyclobutane pyrimidine dimers and 6–4 photoproducts between adjacent pyrimidines; ionizing radiation and disrupted replication can generate single- or double-strand breaks. (ncbi.nlm.nih.gov)\n\nDNA repair pathways recognize different substrates. Direct reversal restores damaged chemistry without replacing a DNA segment: human MGMT transfers an alkyl group from \\(O^{6}\\)-alkylguanine to its own protein, inactivating itself in the process. Photolyases use light energy to reverse ultraviolet photoproducts in many organisms, but humans lack this repair pathway. In base-excision repair, a DNA glycosylase removes an abnormal base, such as uracil, leaving an abasic site. An endonuclease cuts the backbone, the remaining sugar–phosphate residue is processed, and a DNA polymerase and DNA ligase fill and seal the gap. Different glycosylases provide specificity for different lesions, including oxidized bases. (ncbi.nlm.nih.gov)\n\nNucleotide-excision repair removes a short DNA segment containing a bulky, helix-distorting lesion rather than an individual base. Damage recognition, local unwinding, and incisions on both sides release the damaged segment; synthesis using the intact strand and ligation restore continuity. Recognition can begin through genome-wide surveillance or when a lesion stalls RNA polymerase during transcription. Mismatch repair, by contrast, corrects mispaired bases and small insertion–deletion loops left after replication. It identifies the newly synthesized strand, excises DNA containing the error, and resynthesizes that region. In Escherichia coli, MutS and MutL cooperate with MutH to direct repair toward the temporarily unmethylated daughter strand; this methylation-based discrimination is not universal. (ncbi.nlm.nih.gov)\n\nDouble-strand breaks require restoration of both strands. Homologous-recombination repair uses an intact homologous DNA template, usually the sister chromatid in proliferating mammalian cells, and is therefore favored in S and G2 phases. Resection generates single-stranded ends; RAD51-mediated strand invasion allows synthesis from the intact template. This can accurately restore missing sequence, although outcomes depend on template choice and intermediate processing. Non-homologous end joining does not require an extended homologous template: Ku binds the broken ends, and DNA-PK-associated machinery coordinates processing and ligation by the XRCC4–ligase IV complex. Compatible ends can be rejoined accurately; processing damaged or incompatible ends can produce insertions or deletions, while joining incorrect ends can generate rearrangements. (pubmed.ncbi.nlm.nih.gov)\n\nGenome surveillance couples damage detection to repair and cell-cycle control. ATM kinase responds prominently to double-strand breaks, whereas ATR responds to replication stress and exposed single-stranded DNA. Their signaling networks activate checkpoint proteins that delay DNA synthesis or entry into mitosis. Stabilization of p53 can induce p21, which inhibits cyclin-dependent kinases and helps prevent entry into S phase. Persistent damage can instead produce durable growth arrest or apoptosis, depending on cellular context. These responses limit propagation of damaged genomes but do not themselves replace damaged nucleotides. (ncbi.nlm.nih.gov)\n\nInherited repair defects demonstrate the biological consequences of these pathways. Most forms of xeroderma pigmentosum involve nucleotide-excision-repair defects and markedly increased ultraviolet-associated skin-cancer risk; its POLH-associated variant instead impairs damage bypass. Lynch syndrome commonly results from inherited pathogenic variants in the mismatch-repair genes MLH1, MSH2, MSH6, or PMS2, increasing susceptibility to colorectal, endometrial, and other cancers. Repair-pathway failure increases mutation accumulation, but an individual DNA lesion is not equivalent to a disease-causing mutation. (medlineplus.gov)"
    },
    {
      "heading": "Mutation, inheritance, and evolutionary change",
      "text": "A mutation is a change in DNA sequence; a variant is a sequence difference identified by comparison with another sequence, often a reference genome. Neither term necessarily implies disease. Single-nucleotide substitutions replace one base with another: transitions exchange A with G or C with T, whereas transversions exchange a purine with a pyrimidine. Insertions add nucleotides, and deletions remove them. Repeat expansions increase the number of consecutive copies of a sequence motif. At larger scales, copy-number changes delete or duplicate DNA segments, sometimes encompassing several genes. Chromosome rearrangements include inversions, which reverse a segment’s orientation, and translocations, which move segments between chromosome positions. A balanced rearrangement can change DNA organization without changing its total amount, although its breakpoints may disrupt genes or regulatory relationships. (ncbi.nlm.nih.gov)\n\nIn sexually reproducing animals, germline mutations occur in cells that produce eggs or sperm and can enter the next generation; somatic mutations arise in other cells and ordinarily pass only to their cellular descendants. An inherited variant was transmitted by a parent, whereas a de novo variant is newly detected in an offspring and absent from the parental samples examined. It may have arisen in a parental germ cell or during early embryonic development. These categories therefore describe different aspects of inheritance: a new germline mutation can subsequently become an inherited variant. Mosaicism occurs when cells within one individual have different genetic sequences because a mutation arose after fertilization. Its distribution depends on when and in which cell lineage the change occurred; germline mosaicism can permit transmission even when the variant is absent from a parent’s sampled blood cells. (medlineplus.gov)\n\nMutation rates can be measured by comparing offspring genomes with those of both parents. In 2012, Augustine Kong and colleagues sequenced 78 Icelandic parent–offspring trios and estimated an average de novo single-nucleotide mutation rate of \\(1.20 \\times 10^{-8}\\) per nucleotide per generation, with an average paternal age of 29.7 years. Within that sample, each additional year of paternal age was associated with approximately two additional mutations in the offspring. These are study-specific estimates rather than constants applicable to every organism or family, and they concern newly arising sequence changes, not the probability that any particular mutation causes disease. (nature.com)\n\nA variant’s effect depends on its position and molecular context. Within a protein-coding region, a substitution may be synonymous, leaving the encoded amino acid unchanged; missense, replacing an amino acid; or nonsense, creating a premature stop codon. An insertion or deletion whose length is not divisible by three generally shifts the reading frame. Changes at splice sites can alter RNA splicing, while changes in promoters or enhancers can affect the amount, timing, or cellular location of gene expression. Copy-number changes can alter gene dosage. Repeat expansions provide another mechanism: expansion of the CAG tract in HTT lengthens a glutamine sequence in huntingtin and is associated with Huntington disease. Many other variants have no detectable effect under the conditions examined; even an amino-acid replacement need not appreciably alter protein function. Molecular effect, observable phenotype, and disease association are therefore distinct questions. (medlineplus.gov)\n\nHeritable mutations supply new alleles and contribute to genetic diversity. Natural selection changes their frequencies when they affect reproductive success: purifying selection tends to remove deleterious variants, whereas positive selection can increase advantageous ones. Genetic drift changes frequencies through chance sampling between generations, allowing neutral variants to disappear or become fixed. In an idealized diploid population with constant size \\(N\\), random mating, and no selection, a new neutral allele present in one copy has fixation probability \\(1/(2N)\\). Across longer timescales, accumulated sequence differences provide evidence for molecular phylogenetics: homologous DNA sequences are compared to infer common ancestry and branching relationships. Rates differ among genomic regions and lineages, and a gene’s genealogy need not exactly match the species history, so sequence differences are not a universal, uncalibrated clock. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Recombination and the movement of DNA",
      "text": "Homologous recombination exchanges or copies information between DNA molecules with extensive sequence similarity. In a common double-strand-break pathway, processing of the broken ends exposes single-stranded 3′ tails. RecA in bacteria, or related proteins such as RAD51 in eukaryotes, promotes the search for a homologous sequence and invasion of an intact double helix. The invading strand pairs with its complementary strand while displacing the other, forming a displacement loop, or D-loop. This produces heteroduplex DNA: a duplex whose strands originated from different DNA molecules. The invading 3′ end can then prime DNA synthesis using the intact molecule as a template. Sequence differences between the partners can create mismatches within the heteroduplex; their correction can produce gene conversion, a nonreciprocal transfer of sequence information. (ncbi.nlm.nih.gov)\n\nThe processing of these intermediates determines whether recombination exchanges the surrounding chromosome regions. Capture of the second broken end can produce a double Holliday junction, containing two four-stranded DNA connections. Cleavage and rejoining of junctions can yield either a crossover, with reciprocal exchange of flanking regions, or a non-crossover, without that exchange. Non-crossovers also arise through synthesis-dependent strand annealing: the extended invading strand leaves the template and anneals to the other broken end, after which synthesis and ligation complete repair. Thus, non-crossover recombination can transfer local sequence information without exchanging chromosome arms. Experiments reported in 2007 by McMahill, Sham, and Bishop demonstrated synthesis-dependent strand annealing during budding-yeast meiosis, supporting distinct routes to crossover and non-crossover products. (pubmed.ncbi.nlm.nih.gov)\n\nDuring prophase I of meiosis, recombination occurs between nonsister chromatids of paired homologous chromosomes. In many organisms, the enzyme SPO11 initiates the process by generating programmed double-strand breaks. Crossovers reshuffle maternally and paternally inherited alleles along individual chromatids, rather than merely assorting whole chromosomes. They also have a mechanical role: crossover sites appear cytologically as chiasmata, which, together with sister-chromatid cohesion, maintain connections between homologues until their separation at the first meiotic division. Recombination is not uniformly distributed; hotspots concentrate initiation events, and crossover interference reduces the likelihood of another crossover nearby. Requirements differ among organisms, but in most, failure to establish appropriate crossovers increases chromosome-segregation errors. (ncbi.nlm.nih.gov)\n\nSite-specific recombination instead uses specialized recombinases that recognize short, defined DNA sites. Depending on the arrangement and orientation of those sites, reactions can integrate, excise, or invert DNA. For example, bacteriophage λ integrates its genome into the Escherichia coli chromosome through recombination between phage and bacterial attachment sites. Transposition moves a mobile element to a new location without requiring extensive homology between the element and its destination. DNA transposons can move by cut-and-paste or replicative mechanisms, whereas retrotransposons generate new DNA copies through an RNA intermediate. Programmed V(D)J recombination provides another form of targeted rearrangement: RAG1 and RAG2 recognize recombination signals and cleave DNA during lymphocyte development, while cellular end-joining machinery assembles immune-receptor gene segments. This is mechanistically distinct from conservative recombinase-mediated integration. (ncbi.nlm.nih.gov)\n\nHorizontal gene transfer moves DNA between cells, rather than simply rearranging it within a genome. In transformation, a competent bacterium takes up extracellular DNA, which may subsequently recombine with its chromosome. In transduction, a bacteriophage delivers bacterial DNA: generalized transduction follows packaging of host DNA fragments, whereas specialized transduction transfers particular genes adjoining a prophage integration site after abnormal excision. Conjugation requires cell-to-cell contact and a dedicated transfer apparatus. The E. coli F plasmid, for example, transfers a DNA strand that is converted into double-stranded plasmid DNA in the recipient; when F is integrated into the donor chromosome, adjacent chromosomal DNA can also be transferred. Entry and stable inheritance are separate events: incoming DNA may recombine into the chromosome, persist as an independently replicating plasmid, or be lost. (ncbi.nlm.nih.gov)"
    },
    {
      "heading": "Discovery of DNA and evidence for its biological role",
      "text": "In 1869, Friedrich Miescher isolated a phosphorus-rich substance from the nuclei of white blood cells collected from pus on surgical bandages. Working in Felix Hoppe-Seyler’s laboratory in Tübingen, he called it “nuclein” because of its nuclear origin; his findings were published in 1871. This was a chemical discovery, not a demonstration of hereditary function. Subsequent work separated nucleic acids from associated proteins. Albrecht Kossel and colleagues identified their nitrogenous bases, while Phoebus Levene’s investigations established the sugar–phosphate–base organization of nucleotides and identified deoxyribose in DNA in 1929. Levene’s influential but incorrect tetranucleotide hypothesis proposed a repetitive arrangement of the four bases, making DNA appear less capable of encoding diverse hereditary information than proteins. (pmc.ncbi.nlm.nih.gov)\n\nFrederick Griffith’s 1928 experiments demonstrated the transfer of a biological property without identifying the responsible chemical. He studied Streptococcus pneumoniae: encapsulated, smooth-colony bacteria were virulent in mice, whereas unencapsulated, rough-colony bacteria were comparatively harmless. Neither heat-killed smooth bacteria nor live rough bacteria alone produced the lethal outcome observed when the two were combined. From mice injected with the mixture, Griffith recovered living, encapsulated bacteria. Something from the dead cells had therefore “transformed” the living cells. In 1944, Oswald Avery, Colin MacLeod, and Maclyn McCarty purified the transforming substance from type III pneumococci and showed that it induced a stable, heritable change in capsule production. Protein-digesting enzymes and ribonuclease did not eliminate activity, whereas preparations capable of degrading DNA did. Chemical analyses supported identification of the active material as DNA, although the authors explicitly considered the possibility of undetected contaminants. (genome.gov)\n\nAlfred Hershey and Martha Chase supplied independent evidence in 1952 using the T2 bacteriophage, which infects Escherichia coli. They distinguished its components with radioactive isotopes: \\(^{32}\\mathrm{P}\\) labeled DNA, while \\(^{35}\\mathrm{S}\\) labeled sulfur-containing proteins. After phages attached to bacteria, agitation in a blender detached much of the external protein coat; centrifugation separated bacterial cells from this material. Much of the phosphorus label remained associated with the cells, whereas most sulfur label remained outside. The infected cells could nevertheless produce progeny phages. These results supported a role for DNA, rather than the bulk of the incoming protein coat, in directing phage reproduction; they did not establish that every virus uses DNA as its genetic material. (repository.cshl.edu)\n\nErwin Chargaff’s measurements, reported around 1950, established that DNA base composition differed among species but that adenine and thymine occurred in approximately equal amounts, as did guanine and cytosine: \\(A \\approx T\\) and \\(G \\approx C\\). These observations contradicted a uniform, repetitive tetranucleotide composition and constrained possible structural models. At King’s College London, Maurice Wilkins investigated DNA fibers by X-ray diffraction, while Rosalind Franklin and Raymond Gosling obtained detailed measurements distinguishing the A and B forms under different hydration conditions. Their B-form image known as Photograph 51, obtained in 1952, displayed the characteristic cross-shaped pattern of a helix. Franklin’s analyses also constrained its dimensions and placed the phosphate groups toward the outside. (dnalc.cshl.edu)\n\nJames Watson and Francis Crick combined chemical constraints, Chargaff’s findings, and diffraction evidence from the King’s College researchers in a model of two oppositely directed chains with sugar–phosphate backbones outside and complementary base pairs inside. Their structural proposal appeared in Nature on April 25, 1953, alongside separate experimental papers by Wilkins, Alexander Stokes, and Herbert Wilson, and by Franklin and Gosling. Specific A–T and G–C pairing explained the base ratios and suggested how either strand could specify a new complementary strand. The model thus connected molecular structure with a possible copying mechanism, but did not itself experimentally establish how DNA replicated. (nature.com)\n\nMatthew Meselson and Franklin Stahl tested that mechanism in 1958. They grew E. coli with heavy nitrogen, \\(^{15}\\mathrm{N}\\), then shifted the culture to predominantly \\(^{14}\\mathrm{N}\\) and separated extracted DNA by cesium chloride density-gradient centrifugation. After one generation, DNA formed an intermediate-density band rather than separate heavy and light bands, excluding conservative replication. After approximately two generations, intermediate and light bands appeared, rather than the single progressively lighter band predicted by dispersive replication. The results supported semiconservative DNA replication: each daughter molecule retains one parental strand and contains one newly synthesized strand. This demonstrated the inheritance pattern of DNA’s strands, distinct from both identifying DNA as hereditary material and determining its three-dimensional structure. (phys.ens.psl.eu)"
    },
    {
      "heading": "Isolation, detection, amplification, and sequencing",
      "text": "DNA extraction separates nucleic acids from cellular material while limiting degradation. Disruption and detergent-mediated lysis release DNA; proteases remove associated proteins, and RNase treatment reduces RNA contamination. Purification can use organic extraction followed by alcohol precipitation, or solid-phase binding: silica membranes and silica-coated magnetic particles bind DNA under high-salt conditions, allowing contaminants to be washed away before low-salt elution. Sample preservation affects the recovered material: nucleases, acidic conditions, and mechanical shearing can degrade or fragment DNA. Freezing or suitable preservation media limit deterioration before extraction; gentle handling is particularly important when long molecules are required. (ncbi.nlm.nih.gov)\n\nQuality assessment distinguishes concentration, purity, and integrity. Ultraviolet absorbance at 260 nm estimates nucleic-acid concentration but also detects RNA and some contaminants; DNA-selective fluorescent dyes provide a more specific measurement. An \\(A_{260}/A_{280}\\) ratio near 1.8 is consistent with relatively pure DNA, but depends on pH and does not establish that DNA is intact or free from amplification inhibitors. Gel electrophoresis or capillary-based sizing reveals fragment distributions. Preserving long fragments is important for long-read sequencing, whereas short-target assays can use partially fragmented material. (qiagen.com)\n\nIn agarose gel electrophoresis, negatively charged DNA migrates toward the positive electrode; linear fragments are separated principally by size, with smaller fragments generally moving faster. A size ladder provides a comparison, although circular DNA migration also depends on molecular topology. Sequence-specific detection instead uses labeled probes that hybridize to complementary DNA. Southern blotting, published by Edwin Southern in 1975, combines these principles: separated DNA fragments are denatured, transferred to a membrane, and detected with a probe, identifying the sizes of fragments containing a target sequence. Fluorescence in situ hybridization (FISH) applies fluorescent probes to fixed cells or chromosomes, retaining spatial information and revealing sequence location or copy number. Neither method ordinarily determines a complete nucleotide sequence. (ncbi.nlm.nih.gov)\n\nPolymerase chain reaction (PCR) selectively amplifies a region bounded by two primers through repeated denaturation, primer annealing, and extension by a DNA polymerase. Thermostable polymerases made repeated thermal cycling practical; Randall Saiki and colleagues demonstrated this approach in 1988. Conventional PCR typically examines the final product, often as a gel band, and is not a reliable measure of starting quantity. Quantitative PCR (qPCR) monitors fluorescence during amplification: the threshold cycle, commonly denoted \\(C_q\\) or \\(C_t\\), supports quantification against standards or reference samples. Digital PCR partitions a sample into many reactions and scores endpoint fluorescence as positive or negative. Assuming random partitioning, the mean target occupancy is \\(\\lambda=-\\ln(1-p)\\), where \\(p\\) is the positive fraction; this corrects for partitions containing multiple targets and enables concentration estimates without a calibration curve. (pubmed.ncbi.nlm.nih.gov)\n\nDNA sequencing determines nucleotide order. Frederick Sanger, Steven Nicklen, and Alan Coulson’s 1977 chain-termination method incorporates dideoxynucleotides lacking the 3′ hydroxyl needed for extension. Automated implementations separate fluorescently labeled products by capillary electrophoresis, yielding reads of several hundred bases. Massively parallel short-read sequencing analyzes large numbers of adapter-bearing fragments simultaneously; sequencing-by-synthesis platforms identify successive nucleotide incorporations. Long-read methods examine individual molecules: single-molecule real-time sequencing observes polymerase activity, while nanopore sequencing infers sequence from electrical-current changes as DNA passes through a pore. Reads extending thousands to hundreds of thousands of bases can span repeats that are difficult to resolve with short reads. Repeated observation of a circular template can improve single-molecule consensus accuracy. (pubmed.ncbi.nlm.nih.gov)\n\nSequence analysis separates base-call accuracy from coverage and reconstruction. A quality score \\(Q=-10\\log_{10}(e)\\) expresses the estimated error probability \\(e\\); Q30 corresponds to a 0.1% estimated error per base. Mean coverage depth is approximately total sequenced bases divided by target length: 30-fold coverage means thirty observations per position on average, not uniform coverage everywhere. Reference-based analysis aligns reads to an existing genome, whereas de novo assembly reconstructs contiguous sequences from overlaps or shared sequence patterns. Repeats, uneven sampling, and ambiguous alignment can therefore limit reconstruction even when individual reads have high accuracy. (illumina.com)"
    },
    {
      "heading": "Recombinant DNA, genome editing, and molecular engineering",
      "text": "Recombinant DNA is constructed by deliberately joining DNA segments in the laboratory; its components need not come from different species. Unlike natural genetic recombination, the term identifies an engineered construct rather than an exchange occurring without laboratory intervention. In restriction-based cloning, restriction endonucleases cleave DNA at particular recognition sequences, producing staggered, “sticky” ends or blunt ends. Compatible sticky ends associate through complementary base pairing, and DNA ligase seals the sugar–phosphate backbone by forming phosphodiester bonds. In 1973, Stanley Cohen, Annie Chang, Herbert Boyer, and Robert Helling demonstrated that restriction-generated fragments from separate plasmids could be joined and introduced into Escherichia coli, where the resulting molecules replicated and retained properties of both parents. (genome.gov)\n\nA cloning vector carries an inserted DNA segment into a host and supports its maintenance or replication. A typical plasmid vector contains a replication origin, a selectable marker, and sites for inserting DNA; an expression vector additionally supplies regulatory sequences appropriate for producing RNA or protein. Following introduction into bacteria, individual colonies provide populations carrying particular constructs: this is molecular cloning, not cloning an entire organism. Selection identifies cells retaining a vector but does not, by itself, establish that the desired insert is present or correctly oriented. Restriction analysis and PCR can screen candidate clones, while sequencing verifies nucleotide sequence, insert–vector junctions, and reading frames. (genome.gov)\n\nSynthetic DNA can be made from a designed sequence without first isolating that sequence from an organism. Chemical synthesis commonly builds short oligonucleotides through repeated solid-phase phosphoramidite reactions; longer constructs are assembled from overlapping pieces. In the method reported by Daniel Gibson and colleagues in 2009, an exonuclease exposes complementary single-stranded ends, overlapping fragments anneal, a DNA polymerase fills gaps, and ligase seals remaining nicks. This enables assembly without requiring a restriction site at every junction. Synthesis and assembly do not guarantee sequence fidelity: the finished construct is checked against the design by sequencing, because a fragment of the expected size can still contain substitutions, missing bases, or incorrect junctions. (nature.com)\n\nCRISPR-associated genome editing changes DNA at selected genomic locations rather than merely constructing an independently maintained plasmid. Martin Jinek and colleagues, including Jennifer Doudna and Emmanuelle Charpentier, demonstrated programmable RNA-guided cleavage by Cas9 in 2012. A guide RNA directs Cas9 through complementarity to a target adjacent to a suitable protospacer-adjacent motif, or PAM. In nuclease-based editing, the resulting DNA break becomes a substrate for cellular DNA repair: end joining can generate insertions or deletions that disrupt a gene, whereas template-directed repair can incorporate a specified change from donor DNA. Studies in 2013 demonstrated these approaches in mammalian cells. Target recognition therefore does not guarantee a particular editing outcome; sequencing establishes which changes occurred. Editing changes sequence, sequencing reads it, and gene-expression measurements assess its downstream activity. (pmc.ncbi.nlm.nih.gov)\n\nRecombinant constructs support research into gene function and established industrial protein production. Human insulin provides an early example: work at Genentech and City of Hope in 1978 used synthetic genes encoding its A and B chains. The experimental report, published in January 1979, described separate expression in E. coli as fusion proteins, release and purification of the peptides, and their combination under conditions permitting disulfide-bond formation. On October 28, 1982, the US Food and Drug Administration approved Humulin, the first approved medical product derived from recombinant DNA technology. The engineered DNA directs cellular manufacture of the protein; it is not itself the insulin administered. (pubmed.ncbi.nlm.nih.gov)\n\nDNA also serves as an engineering material independently of protein coding. In 2006, Paul Rothemund demonstrated DNA origami, using more than 200 short “staple” strands to fold a roughly 7-kilobase scaffold into structures approximately 100 nanometres across through molecular self-assembly. Experimental DNA data storage instead encodes digital information in nucleotide sequences, writes it through synthesis, and recovers it through sequencing and decoding. Yaniv Erlich and Dina Zielinski’s 2017 DNA Fountain experiment demonstrated error-free recovery using a coding scheme designed to tolerate missing molecules and sequence constraints. These demonstrations establish structural programmability and information storage, respectively, rather than the mature manufacturing role exemplified by recombinant insulin. (nature.com)"
    },
    {
      "heading": "DNA evidence in medicine, identification, and environmental research",
      "text": "In inherited-disease investigation, DNA sequencing can identify variants in individual genes, gene panels, or broader genomic regions. Interpretation combines the molecular result with symptoms, family history, inheritance patterns, and evidence about the variant’s effects. For example, pathogenic variants in BRCA1 or BRCA2 can establish an inherited predisposition to particular cancers, but do not establish that cancer will develop. A variant of uncertain significance is a detected sequence difference whose clinical meaning remains unresolved; an uninformative negative result does not exclude every genetic cause. Sequencing also supports investigation of rare disorders, as in the US National Institutes of Health’s Undiagnosed Diseases Program. (cancer.gov)\n\nTumor characterization examines acquired DNA changes that may distinguish cancer subtypes or serve as treatment-related biomarkers. Comparing tumor DNA with non-tumor DNA helps distinguish somatic changes from germline variants; a tumor result alone is not equivalent to a finding of inherited susceptibility. Clinical interpretation also depends on cancer type and the evidence connecting a particular alteration to treatment response. In infectious-disease investigation, sequencing identifies and compares pathogen genomes. The Centers for Disease Control and Prevention’s Listeria whole-genome sequencing project compares isolates from patients, foods, and environmental samples to investigate outbreaks. Genomic similarity supports a possible connection, but food-exposure histories and other epidemiological evidence remain necessary to establish an outbreak source. (cancer.gov)\n\nForensic profiling commonly measures short tandem repeats (STRs): short sequence motifs whose repeat counts vary among individuals. A profile records alleles at selected loci rather than an entire genome; at autosomal loci, one allele is inherited from each parent. On January 1, 2017, the US Federal Bureau of Investigation expanded the CODIS core set from 13 to 20 loci. Kinship analysis evaluates whether allele-sharing patterns better support a specified relationship, such as parent–child, than an alternative relationship. This supports parentage testing and identification of missing people or human remains, while allowing for mutation and uncertainty in family relationships. (nist.gov)\n\nThe statistical weight of a forensic comparison is often expressed as a likelihood ratio, \\(LR=P(E\\mid H_1)/P(E\\mid H_2)\\), comparing the probability of the observed evidence under two stated propositions. It is not the probability that a suspect is guilty, nor the probability that a proposed contributor supplied the DNA. Mixed samples require consideration of contributor number, degradation, missing alleles, and laboratory artifacts. A profile match concerns possible source identity: it does not by itself establish when DNA was deposited, whether deposition was direct or involved secondary transfer, or what activity produced the trace. Validation must therefore address the kinds of samples and interpretations for which a method is used. (nvlpubs.nist.gov)\n\nAncient DNA recovered from bones, teeth, and other preserved material provides evidence about past populations and evolutionary relationships. In 2010, David Reich and colleagues reported a nuclear genome from a finger bone in Denisova Cave, Siberia, identifying an archaic population genetically distinct from Neanderthals and evidence of its contribution to present-day Melanesian ancestry. Ancient molecules are typically fragmented and chemically damaged; cytosine deamination can produce apparent cytosine-to-thymine substitutions. Modern human and microbial contamination can overwhelm endogenous DNA, making laboratory precautions, contamination estimates, and damage-pattern analysis central to authentication. Environmental DNA (eDNA), by contrast, samples genetic material released into water, soil, or other surroundings. US Geological Survey laboratories use it to detect organisms including invasive carp and rare mussels. A detection establishes the presence of genetic material, not necessarily a living organism at the sampling point; transport, persistence, contamination, and incomplete sampling complicate both positive and negative results and estimates of abundance. (nature.com)\n\nHuman genomic studies also require explicit treatment of informed consent and genetic-data privacy. Consent processes distinguish the purpose of testing, future sample use, data sharing, and whether individual or incidental findings will be returned. Genomic records can reveal information relevant to relatives, and removing names does not eliminate re-identification risks when sequence data are combined with other information. Confidentiality arrangements and access restrictions therefore form part of study design, alongside explanations of uncertain findings and the limits of the analysis. (genome.gov)"
    },
    {
      "heading": "Origins of DNA and unresolved research questions",
      "text": "The RNA world hypothesis proposes that early evolving systems used RNA both to carry hereditary information and to catalyse reactions, before DNA became the principal genomic material. A proposed intermediate is an RNA–protein system in which emerging enzymes produced DNA precursors and copied RNA information into DNA. Modern biochemistry establishes that such transformations are chemically possible: ribonucleotide reductases convert ribonucleotides into deoxyribonucleotides, thymidylate synthases produce thymidylate from deoxyuridylate, and reverse transcriptases synthesize DNA using RNA templates. These mechanisms do not establish the historical order in which they arose. DNA’s resistance to cleavage involving the sugar’s 2′-hydroxyl group, and the use of thymine to distinguish ordinary bases from uracil generated by cytosine deamination, provide plausible advantages for hereditary storage rather than direct evidence of its origin. (ncbi.nlm.nih.gov)\n\nOther hypotheses allow RNA and DNA components to have emerged together rather than through a strictly RNA-first sequence. In 2020, Jianfeng Xu and colleagues reported a laboratory pathway producing RNA pyrimidine and DNA purine nucleosides under proposed prebiotic conditions, supporting the chemical plausibility of mixed precursor pools. Such experiments establish reactions under specified conditions; they do not demonstrate that those reactions occurred on early Earth or produced a self-replicating organism. Major unresolved steps include connecting precursor synthesis to polymer formation, achieving sufficiently accurate template copying, and explaining the transition from mixed nucleic-acid systems to separate DNA-storage and RNA-expression roles. (nature.com)\n\nFor non-coding DNA, a central difficulty is specifying what “function” means. A 2014 analysis by Manolis Kellis and colleagues distinguished biochemical activity, consequences of genetic perturbation, and evolutionary evidence of selection. Transcription or transcription-factor binding identifies molecular activity, but does not alone demonstrate an effect on organismal survival or reproduction. Conversely, conservation under purifying selection supports functional constraint, although recently evolved or lineage-specific elements may lack deep conservation. Deleting a candidate enhancer can reveal a regulatory requirement, but a negative result may reflect redundant elements, an untested developmental stage, or an environmental condition absent from the experiment. Assigning function therefore requires a defined outcome and biological context, not simply a genome-wide activity label. (pmc.ncbi.nlm.nih.gov)\n\nResearch on DNA structure likewise distinguishes sequence potential from structures actually formed in cells. Guanine-rich sequences can form G-quadruplexes, but their folding depends on accessibility and molecular surroundings. In 2021, Sara Lago and colleagues compared liposarcoma cells with keratinocytes and found that identical promoter sequences could have different G-quadruplex-folding states; folded promoters were associated with open chromatin, high transcription, and binding by AP-1 and SP1. This illustrates why sequence alone cannot specify a regulatory state. An unresolved issue is the direction of causation: whether a structure recruits regulatory proteins, whether transcription and chromatin opening permit its formation, or whether both processes reinforce each other. Association must therefore be distinguished from a demonstrated mechanism at a particular locus. (nature.com)\n\nAssigning effects to individual variants presents a related gap between molecular measurement and biological consequence. A 2021 SNP-SELEX study tested 270 human transcription factors against 95,886 non-coding variants, generating approximately 828 million protein–DNA interaction measurements. Such assays can identify allele-dependent binding in vitro, but altered affinity does not by itself establish altered expression in a particular tissue or an effect on a whole-organism phenotype. The genomic location, available regulatory proteins, and cellular environment remain essential to interpretation. Consequently, detecting a variant and measuring one molecular effect are distinct from establishing its causal contribution to a biological trait. (nature.com)\n\nComplex genomes also impose reconstruction limits. Nearly identical repeats can obscure copy number, order, and the assignment of sequences to parental chromosomes. In 2022, Sergey Nurk and colleagues combined complementary long-read technologies to produce T2T-CHM13, resolving approximately 200 million previously unavailable base pairs, including centromeric regions; its nearly homozygous source simplified assembly but did not represent all human variation. The 2023 draft human pangenome incorporated 47 phased diploid assemblies to represent alternative sequences and reduce dependence on a single linear reference. These resources improve sequence detection and reconstruction, while leaving the experimental task of determining what newly resolved sequences do in particular cells and organisms. (eichlerlab.gs.washington.edu)"
    }
  ],
  "links": [
    "nucleic-acid",
    "polymer",
    "nucleotide",
    "gene",
    "dna-replication",
    "transcription-biology",
    "genome",
    "molecular-biology",
    "ribose",
    "phosphodiester-bond",
    "covalent-bond",
    "rna",
    "hydrolysis",
    "hydrogen-bond",
    "protein",
    "molecular-recognition",
    "base-pair",
    "x-ray-crystallography",
    "g-quadruplex",
    "ion",
    "gene-expression",
    "dna-supercoiling",
    "ph",
    "temperature",
    "dna-melting-temperature",
    "thermodynamics",
    "nucleic-acid-hybridization",
    "bacteria",
    "archaea",
    "eukaryote",
    "cell-nucleus",
    "chromosome",
    "prokaryote",
    "mitochondrion",
    "chloroplast",
    "transfer-rna",
    "ribosomal-rna",
    "endosymbiotic-theory",
    "plasmid",
    "horizontal-gene-transfer",
    "virus",
    "chromatin",
    "nucleosome",
    "histone",
    "mitosis",
    "cell-cycle",
    "centromere",
    "kinetochore",
    "telomere",
    "genetic-locus",
    "allele",
    "messenger-rna",
    "exon",
    "intron",
    "rna-splicing",
    "promoter-biology",
    "enhancer-biology",
    "transposable-element",
    "natural-selection",
    "dna-polymerase",
    "dna-ligase",
    "dna-proofreading",
    "dna-mismatch-repair",
    "telomerase",
    "rna-polymerase",
    "transcription-factor",
    "translation-biology",
    "ribosome",
    "codon",
    "genetic-code",
    "central-dogma",
    "reverse-transcription",
    "restriction-enzyme",
    "dna-methylation",
    "cell-division",
    "epigenetics",
    "genomic-imprinting",
    "mutation",
    "dna-repair",
    "base-excision-repair",
    "nucleotide-excision-repair",
    "homologous-recombination",
    "non-homologous-end-joining",
    "apoptosis",
    "cancer",
    "mosaicism",
    "amino-acid",
    "reading-frame",
    "phenotype",
    "genetic-diversity",
    "genetic-drift",
    "phylogenetics",
    "chromosomal-crossover",
    "meiosis",
    "homologous-chromosome",
    "vdj-recombination",
    "bacteriophage",
    "isotope",
    "diffraction",
    "gel-electrophoresis",
    "southern-blot",
    "fluorescence-in-situ-hybridization",
    "polymerase-chain-reaction",
    "dna-sequencing",
    "recombinant-dna",
    "crispr",
    "genome-editing",
    "cas-nine",
    "insulin",
    "dna-origami",
    "molecular-self-assembly",
    "dna-data-storage",
    "biomarker",
    "short-tandem-repeat",
    "ancient-dna",
    "environmental-dna",
    "informed-consent",
    "data-privacy",
    "rna-world",
    "enzyme",
    "pangenome"
  ],
  "sources": [
    {
      "title": "The Human Genome - Genomes - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21134/?report=reader"
    },
    {
      "title": "The Molecular Composition of Cells - The Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK9879/"
    },
    {
      "title": "The Structure and Function of DNA - Molecular Biology of the Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26821/"
    },
    {
      "title": "The Human Genome - Genomes - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21134/"
    },
    {
      "title": "Alkaline hydrolysis to remove potentially infectious viral RNA contaminants from DNA - PubMed",
      "url": "https://pubmed.ncbi.nlm.nih.gov/27260412/"
    },
    {
      "title": "Ribonucleic Acid (RNA) Fact Sheet",
      "url": "https://www.genome.gov/about-genomics/educational-resources/fact-sheets/ribonucleic-acid-fact-sheet"
    },
    {
      "title": "Probe",
      "url": "https://www.genome.gov/genetics-glossary/Probe"
    },
    {
      "title": "Biochemistry, DNA Structure - StatPearls - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK538241/"
    },
    {
      "title": "DNA: Alternative Conformations and Biology - Madame Curie Bioscience Database - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK6545/?report=reader"
    },
    {
      "title": "DNA: Alternative Conformations and Biology - Madame Curie Bioscience Database - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK6545/?report=printable"
    },
    {
      "title": "Base-stacking and base-pairing contributions into thermal stability of the DNA double helix - PMC",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC1360284/"
    },
    {
      "title": "The role of DNA shape in protein-DNA recognition - PMC",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2793086/"
    },
    {
      "title": "Molecular structure of a left-handed double helical DNA fragment at atomic resolution",
      "url": "https://www.nature.com/articles/282680a0"
    },
    {
      "title": "DNA: Alternative Conformations and Biology",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK6545/?report=classic"
    },
    {
      "title": "DNA H form requires a homopurine-homopyrimidine mirror repeat",
      "url": "https://pubmed.ncbi.nlm.nih.gov/2825028/"
    },
    {
      "title": "Quantitative Visualization of DNA G-quadruplex Structures in Human Cells",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3622242/"
    },
    {
      "title": "G-quadruplex structures are stable and detectable in human genomic DNA",
      "url": "https://www.nature.com/articles/ncomms2792"
    },
    {
      "title": "Linking topology of large DNA molecules",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2896270/"
    },
    {
      "title": "Characterization of the geometry and topology of DNA pictured as a discrete collection of atoms",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4004093/"
    },
    {
      "title": "Topoisomerase Biology",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK14039/"
    },
    {
      "title": "DNA Replication Mechanisms",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26850/%26lang%3Den/"
    },
    {
      "title": "Topoisomerase IV, not gyrase, decatenates products of site-specific recombination in Escherichia coli",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC316566/"
    },
    {
      "title": "Quantitative and comprehensive decomposition of the ion atmosphere around nucleic acids",
      "url": "https://pubmed.ncbi.nlm.nih.gov/17990882/"
    },
    {
      "title": "A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC19045/"
    },
    {
      "title": "Modeling Formamide Denaturation of Probe-Target Hybrids for Improved Microarray Probe Design in Microbial Diagnostics",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3428302/"
    },
    {
      "title": "Determination of the base composition of deoxyribonucleic acid from its thermal denaturation temperature",
      "url": "https://computingbiology.github.io/docs/marmur1962.pdf"
    },
    {
      "title": "Mechanism of thermal renaturation and hybridization of nucleic acids: Kramers process and universality in Watson-Crick base pairing",
      "url": "https://arxiv.org/abs/0808.1233"
    },
    {
      "title": "Characterization of denaturation and renaturation of DNA for DNA hybridization",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC4168728/"
    },
    {
      "title": "Evaluating digital PCR for the quantification of human nuclear DNA: determining target strandedness",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7931668/"
    },
    {
      "title": "Limitations of methods for measuring the concentration of human genomic DNA and oligonucleotide samples",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6157598/"
    },
    {
      "title": "Genome Anatomies - Genomes - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21120/"
    },
    {
      "title": "Sequence and organization of the human mitochondrial genome - PubMed",
      "url": "https://pubmed.ncbi.nlm.nih.gov/7219534/?dopt=Abstract"
    },
    {
      "title": "Mitochondrial DNA: MedlinePlus Genetics",
      "url": "https://medlineplus.gov/genetics/chromosome/mitochondrial-dna/"
    },
    {
      "title": "Genetic Information in Eucaryotes - Molecular Biology of the Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26909/"
    },
    {
      "title": "Chloroplasts and Other Plastids - The Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK9905/"
    },
    {
      "title": "Most chloroplast DNA of maize seedlings in linear molecules with defined ends and branched forms - PubMed",
      "url": "https://pubmed.ncbi.nlm.nih.gov/14698291/"
    },
    {
      "title": "Studying DNA - Genomes - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21129/"
    },
    {
      "title": "Salmonella typhimurium LT2 mutation affecting the deletion of resistance determinants on R plasmids - PubMed",
      "url": "https://pubmed.ncbi.nlm.nih.gov/6989800/"
    },
    {
      "title": "Genetics - Medical Microbiology - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK7908/?report=classic"
    },
    {
      "title": "Family: Adenoviridae - ICTV",
      "url": "https://ictv.global/report/chapter/adenoviridae/adenoviridae"
    },
    {
      "title": "Family: Parvoviridae - ICTV",
      "url": "https://ictv.global/report/chapter/parvoviridae/parvoviridae"
    },
    {
      "title": "Chromosomes Fact Sheet",
      "url": "https://www.genome.gov/about-genomics/fact-sheets/Chromosomes-Fact-Sheet"
    },
    {
      "title": "Chromatid",
      "url": "https://www.genome.gov/genetics-glossary/Chromatid"
    },
    {
      "title": "Chromosomal DNA and Its Packaging in the Chromatin Fiber",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26834/"
    },
    {
      "title": "ChromEMT: Visualizing 3D chromatin structure and compaction in interphase and mitotic cells",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5646685/"
    },
    {
      "title": "DNA loop extrusion by human cohesin",
      "url": "https://pubmed.ncbi.nlm.nih.gov/31753851/"
    },
    {
      "title": "A pathway for mitotic chromosome formation",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5924687/"
    },
    {
      "title": "Structure-based analysis of HU-DNA binding",
      "url": "https://pubmed.ncbi.nlm.nih.gov/17097674/"
    },
    {
      "title": "Dual architectural roles of HU: formation of flexible hinges and rigid filaments",
      "url": "https://pubmed.ncbi.nlm.nih.gov/15118104/"
    },
    {
      "title": "Structure of histone-based chromatin in Archaea",
      "url": "https://pubmed.ncbi.nlm.nih.gov/28798133/"
    },
    {
      "title": "Histone diversity in the archaeal domain of life",
      "url": "https://pubmed.ncbi.nlm.nih.gov/41986316/"
    },
    {
      "title": "Centromere assembly requires the direct recognition of CENP-A nucleosomes by CENP-N",
      "url": "https://cmgm-new.stanford.edu/biochem/straightlab/pdfs/Centromere%20assembly%20requires%20the%20direct%20recognition%20of.pdf"
    },
    {
      "title": "Shelterin: the protein complex that shapes and safeguards human telomeres",
      "url": "https://pubmed.ncbi.nlm.nih.gov/16166375/"
    },
    {
      "title": "Talking Glossary of Genetic Terms",
      "url": "https://www.genome.gov/node/41621"
    },
    {
      "title": "Genome",
      "url": "https://www.genome.gov/genetics-glossary/Genome"
    },
    {
      "title": "Locus",
      "url": "https://www.genome.gov/genetics-glossary/Locus"
    },
    {
      "title": "Allele",
      "url": "https://www.genome.gov/genetics-glossary/Allele?id=4"
    },
    {
      "title": "Gene",
      "url": "https://www.genome.gov/genetics-glossary/Gene"
    },
    {
      "title": "Ribonucleic Acid (RNA)",
      "url": "https://www.genome.gov/genetics-glossary/Ribonucleic-Acid-RNA"
    },
    {
      "title": "Exon",
      "url": "https://www.genome.gov/genetics-glossary/Exon?id=61"
    },
    {
      "title": "Intron",
      "url": "https://www.genome.gov/genetics-glossary/Intron"
    },
    {
      "title": "Regulation of eukaryotic gene expression by the untranslated gene regions and other non-coding elements",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3474909/"
    },
    {
      "title": "Promoter",
      "url": "https://www.genome.gov/genetics-glossary/Promoter"
    },
    {
      "title": "Intergenic Regions",
      "url": "https://www.genome.gov/genetics-glossary/Intergenic-Regions"
    },
    {
      "title": "Initial sequencing and analysis of the human genome",
      "url": "https://www.nature.com/articles/35057062"
    },
    {
      "title": "DNA Replication Mechanisms",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26850/"
    },
    {
      "title": "How the Eukaryotic Replisome Achieves Rapid and Efficient DNA Replication",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC5222725/"
    },
    {
      "title": "Genome Replication",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21113/"
    },
    {
      "title": "PCNA acts as a stationary loading platform for transiently interacting Okazaki fragment maturation proteins",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC1156965/"
    },
    {
      "title": "The Initiation and Completion of DNA Replication in Chromosomes",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26826/"
    },
    {
      "title": "Topoisomerase II Is Crucial for Fork Convergence during Vertebrate Replication Termination",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC6919565/"
    },
    {
      "title": "Structures of telomerase at several steps of telomere repeat synthesis",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC8857963/"
    },
    {
      "title": "How Telomeres Solve the End-Protection Problem",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2819049/"
    },
    {
      "title": "From DNA to RNA - Molecular Biology of the Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26887/"
    },
    {
      "title": "Transcription in Prokaryotes - The Cell - NCBI Bookshelf",
      "url": "https://ncbi.nlm.nih.gov/books/NBK9850/"
    },
    {
      "title": "Eukaryotic RNA Polymerases and General Transcription Factors - The Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK9935/"
    },
    {
      "title": "Human 5′ → 3′ exonuclease Xrn2 promotes transcription termination at co-transcriptional cleavage sites",
      "url": "https://www.nature.com/articles/nature03035"
    },
    {
      "title": "From RNA to Protein - Molecular Biology of the Cell - NCBI Bookshelf",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26829/"
    },
    {
      "title": "Central Dogma of Molecular Biology",
      "url": "https://www.dna.caltech.edu/courses/cs191/paperscs191/CrickCentralDogma1970.pdf"
    },
    {
      "title": "Viral RNA-dependent DNA Polymerase: RNA-dependent DNA Polymerase in Virions of RNA Tumour Viruses",
      "url": "https://www.nature.com/articles/2261209a0"
    },
    {
      "title": "RNA-dependent DNA polymerase in virions of Rous sarcoma virus",
      "url": "https://pubmed.ncbi.nlm.nih.gov/4316301/"
    },
    {
      "title": "The role of DNA shape in protein-DNA recognition",
      "url": "https://rohslab.usc.edu/Papers/Rohs_etal_Nature.pdf"
    },
    {
      "title": "The mechanism by which DNA adenine methylase and PapI activate the pap epigenetic switch",
      "url": "https://pubmed.ncbi.nlm.nih.gov/14580345/"
    },
    {
      "title": "Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex",
      "url": "https://www.nature.com/articles/30764"
    },
    {
      "title": "Acetylation Mimics Within a Single Nucleosome Alter Local DNA Accessibility In Compacted Nucleosome Arrays",
      "url": "https://www.nature.com/articles/srep34808"
    },
    {
      "title": "Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins",
      "url": "https://www.nature.com/articles/35065132"
    },
    {
      "title": "Human DNA methylomes at base resolution show widespread epigenomic differences",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2857523/"
    },
    {
      "title": "UHRF1 plays a role in maintaining DNA methylation in mammalian cells",
      "url": "https://pubmed.ncbi.nlm.nih.gov/17673620/"
    },
    {
      "title": "Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene",
      "url": "https://www.nature.com/articles/35013100"
    },
    {
      "title": "DNA Repair — Molecular Biology of the Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26879/"
    },
    {
      "title": "Mutation, Repair and Recombination — Genomes",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21114/"
    },
    {
      "title": "DNA Repair — The Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK9900/"
    },
    {
      "title": "Cloning and characterization of a mammalian 8-oxoguanine DNA glycosylase",
      "url": "https://pubmed.ncbi.nlm.nih.gov/9207108/"
    },
    {
      "title": "Origin, Recognition, Signaling and Repair of DNA Double-Strand Breaks in Mammalian Cells",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK6555/?report=reader"
    },
    {
      "title": "Homologous Recombination Deficiencies and Hereditary Tumors",
      "url": "https://pubmed.ncbi.nlm.nih.gov/35008774/"
    },
    {
      "title": "HDR through Homologous Recombination (HRR) or Single Strand Annealing (SSA)",
      "url": "https://pubchem.ncbi.nlm.nih.gov/pathway/Reactome%3AR-HSA-5693567"
    },
    {
      "title": "Biochemical evidence for Ku-independent backup pathways of NHEJ",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC203313/"
    },
    {
      "title": "Intracellular Control of Cell-Cycle Events",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26856/"
    },
    {
      "title": "Proliferation — Holland-Frei Cancer Medicine",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK13035/"
    },
    {
      "title": "Biologic Basis of Radiation Therapy",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK12902/"
    },
    {
      "title": "Xeroderma pigmentosum — MedlinePlus Genetics",
      "url": "https://medlineplus.gov/genetics/condition/xeroderma-pigmentosum/"
    },
    {
      "title": "What is a gene variant and how do variants occur?",
      "url": "https://medlineplus.gov/genetics/understanding/mutationsanddisorders/genemutation/"
    },
    {
      "title": "Human Genomic Variation",
      "url": "https://www.genome.gov/about-genomics/educational-resources/fact-sheets/human-genomic-variation"
    },
    {
      "title": "Rate of de novo mutations and the importance of father’s age to disease risk",
      "url": "https://www.nature.com/articles/nature11396"
    },
    {
      "title": "What kinds of gene variants are possible?",
      "url": "https://medlineplus.gov/genetics/understanding/mutationsanddisorders/possiblemutations/"
    },
    {
      "title": "HTT gene: MedlinePlus Genetics",
      "url": "https://medlineplus.gov/genetics/gene/htt/"
    },
    {
      "title": "How Genomes Evolve - Molecular Biology of the Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26836/"
    },
    {
      "title": "Molecular Phylogenetics - Genomes",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK21122/?report=classic"
    },
    {
      "title": "General Recombination - Molecular Biology of the Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26898/"
    },
    {
      "title": "Synthesis-dependent strand annealing in meiosis",
      "url": "https://pubmed.ncbi.nlm.nih.gov/17988174/"
    },
    {
      "title": "Control of Meiotic Crossovers: From Double-Strand Break Formation to Designation",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/5319444/"
    },
    {
      "title": "Meiosis - Molecular Biology of the Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26840/"
    },
    {
      "title": "Meiosis - WormBook",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK430708/"
    },
    {
      "title": "DNA Rearrangements - The Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK9937/"
    },
    {
      "title": "Site-Specific Recombination - Molecular Biology of the Cell",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK26845/"
    },
    {
      "title": "The DNA Damage Response: Implications on Cancer Formation and Treatment",
      "url": "https://www.ncbi.nlm.nih.gov/books/n/spr9783319601922/pdf/"
    },
    {
      "title": "Genetics - Medical Microbiology",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK7908/"
    },
    {
      "title": "Before Watson and Crick in 1953 Came Friedrich Miescher in 1869",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7268995/"
    },
    {
      "title": "Some Observations on the Early History of Nucleic Acid Chemistry",
      "url": "https://digirepo.nlm.nih.gov/ext/document/101584575X14/PDF/101584575X14.pdf"
    },
    {
      "title": "Animation 15: DNA and proteins are key molecules of the cell nucleus.",
      "url": "https://dnalc.cshl.edu/view/16342-Animation-15-DNA-and-proteins-are-key-molecules-of-the-cell-nucleus-.html"
    },
    {
      "title": "1944: DNA is \"Transforming Principle\"",
      "url": "https://www.genome.gov/25520250/online-education-kit-1944-dna-is-transforming-principle"
    },
    {
      "title": "Studies on the Chemical Nature of the Substance Inducing Transformation of Pneumococcal Types",
      "url": "https://njc.rockefeller.edu/pdf4/Class3-AveryMacLeodMcCarty1944.pdf"
    },
    {
      "title": "Independent Functions of Viral Protein and Nucleic Acid in Growth of Bacteriophage",
      "url": "https://www.vakalis.edu.gr/sites/default/files/files/blogpost/hershey-and-chase-experiment.pdf"
    },
    {
      "title": "Chargaff's ratios, 3D animation with narration",
      "url": "https://dnalc.cshl.edu/view/15495-Chargaff"
    },
    {
      "title": "Chemical Nature and Biological Specificity of the Genetic Material",
      "url": "https://www.govinfo.gov/content/pkg/CPRT-93HPRT42772O/pdf/CPRT-93HPRT42772O.pdf"
    },
    {
      "title": "What Rosalind Franklin truly contributed to the discovery of DNA’s structure",
      "url": "https://www.nature.com/articles/d41586-023-01313-5"
    },
    {
      "title": "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid",
      "url": "https://www.nature.com/articles/171737a0"
    },
    {
      "title": "Molecular structure of deoxypentose nucleic acids",
      "url": "https://pubmed.ncbi.nlm.nih.gov/13054693/"
    },
    {
      "title": "DNA extraction technologies",
      "url": "https://www.qiagen.com/us/knowledge-and-support/knowledge-hub/bench-guide/dna/introduction/dna-extraction-technologies"
    },
    {
      "title": "DNA sample storage: best practices after extraction",
      "url": "https://www.qiagen.com/us/knowledge-and-support/knowledge-hub/bench-guide/dna/handling-dna/storage-of-dna"
    },
    {
      "title": "Sample storage prior to extraction of genomic DNA",
      "url": "https://www.qiagen.com/ch/knowledge-and-support/knowledge-hub/bench-guide/dna/handling-dna/sample-storage-prior-to-extraction-of-genomic-dna"
    },
    {
      "title": "DNA Quantification Methods and Applications",
      "url": "https://www.qiagen.com/cn/knowledge-and-support/knowledge-hub/bench-guide/dna/analysing-dna/quantification-of-dna"
    },
    {
      "title": "Detection of specific sequences among DNA fragments separated by gel electrophoresis",
      "url": "https://pubmed.ncbi.nlm.nih.gov/1195397/"
    },
    {
      "title": "Detection of specific DNA sequences—the southern transfer",
      "url": "https://pubmed.ncbi.nlm.nih.gov/21374173/"
    },
    {
      "title": "Fluorescence In Situ Hybridization Fact Sheet",
      "url": "https://www.genome.gov/about-genomics/fact-sheets/Fluorescence-In-Situ-Hybridization"
    },
    {
      "title": "Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase",
      "url": "https://pubmed.ncbi.nlm.nih.gov/2448875/?dopt=Abstract"
    },
    {
      "title": "Introduction to Digital PCR",
      "url": "https://www.bio-rad.com/de/life-science/learning-center/introduction-to-digital-pcr"
    },
    {
      "title": "Absolute vs. Relative Quantification for qPCR",
      "url": "https://www.thermofisher.com/us/en/home/life-science/pcr/real-time-pcr/real-time-pcr-learning-center/real-time-pcr-basics/absolute-vs-relative-quantification-real-time-pcr.html"
    },
    {
      "title": "Droplet Digital PCR Applications Guide",
      "url": "https://www.bio-rad.com/webroot/web/pdf/lsr/literature/Bulletin_6407.pdf"
    },
    {
      "title": "Recombinant DNA Technology",
      "url": "https://www.genome.gov/genetics-glossary/Recombinant-DNA-Technology"
    },
    {
      "title": "Construction of Biologically Functional Bacterial Plasmids In Vitro",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC427208/?page=1"
    },
    {
      "title": "Cloning",
      "url": "https://www.genome.gov/genetics-glossary/Cloning"
    },
    {
      "title": "Molecular Cloning Technical Guide",
      "url": "https://www.neb.com/en/-/media/nebus/files/brochures/molcloning_tech_guide.pdf"
    },
    {
      "title": "Synthesis and site-specific incorporation of a simple fluorescent pyrimidine",
      "url": "https://www.nature.com/articles/nprot.2006.464"
    },
    {
      "title": "Enzymatic assembly of DNA molecules up to several hundred kilobases",
      "url": "https://static.igem.org/mediawiki/2013/d/d0/3.1-_Enzymatic_Assembly_of_DNA_Molecules_up_to_Several_Hundred_Kilobases.pdf"
    },
    {
      "title": "A programmable dual RNA-guided DNA endonuclease in adaptive bacterial immunity",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/6286148/"
    },
    {
      "title": "Multiplex Genome Engineering Using CRISPR/Cas Systems",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/3795411/"
    },
    {
      "title": "Expression in Escherichia coli of chemically synthesized genes for human insulin",
      "url": "https://pubmed.ncbi.nlm.nih.gov/85300/"
    },
    {
      "title": "Genetic Testing Fact Sheet",
      "url": "https://www.cancer.gov/about-cancer/causes-prevention/genetics/genetic-testing-fact-sheet"
    },
    {
      "title": "DNA Sequencing Fact Sheet",
      "url": "https://www.genome.gov/about-genomics/fact-sheets/DNA-Sequencing-Fact-Sheet"
    },
    {
      "title": "Biomarker Testing for Cancer Treatment",
      "url": "https://www.cancer.gov/about-cancer/treatment/types/biomarker-testing-cancer-treatment"
    },
    {
      "title": "The Listeria Whole Genome Sequencing Project",
      "url": "https://www.cdc.gov/listeria/surveillance/whole-genome-sequencing.html"
    },
    {
      "title": "DNA profile",
      "url": "https://www.nist.gov/image/dnaprofilejpg"
    },
    {
      "title": "Kinship Information",
      "url": "https://strbase-archive.nist.gov/kinship.htm"
    },
    {
      "title": "Low Level DNA & Kinship Analysis",
      "url": "https://strbase-archive.nist.gov/pub_pres/ForensicsNIST-LTDNA-kinship.pdf"
    },
    {
      "title": "DNA Mixture Interpretation: A NIST Scientific Foundation Review",
      "url": "https://nvlpubs.nist.gov/nistpubs/ir/2024/NIST.IR.8351.pdf"
    },
    {
      "title": "Genetic history of an archaic hominin group from Denisova Cave in Siberia",
      "url": "https://www.nature.com/articles/nature09710"
    },
    {
      "title": "Separating endogenous ancient DNA from modern day contamination in a Siberian Neandertal",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC3926038/"
    },
    {
      "title": "Environmental DNA (eDNA)",
      "url": "https://www.usgs.gov/labs/ohio-water-microbiology-laboratory/environmental-dna-edna"
    },
    {
      "title": "The Interplay between RNA and DNA Modifications: Back to the RNA World",
      "url": "https://www.ncbi.nlm.nih.gov/books/NBK6338/"
    },
    {
      "title": "Selective prebiotic formation of RNA pyrimidine and DNA purine nucleosides",
      "url": "https://www.nature.com/articles/s41586-020-2330-9"
    },
    {
      "title": "Defining functional DNA elements in the human genome",
      "url": "https://pmc.ncbi.nlm.nih.gov/articles/4035993/"
    },
    {
      "title": "Promoter G-quadruplexes and transcription factors cooperate to shape the cell type-specific transcriptome",
      "url": "https://www.nature.com/articles/s41467-021-24198-2"
    },
    {
      "title": "Systematic analysis of binding of transcription factors to noncoding variants",
      "url": "https://www.nature.com/articles/s41586-021-03211-0"
    },
    {
      "title": "The complete sequence of a human genome",
      "url": "https://eichlerlab.gs.washington.edu/help/glogsdon/Website/Nurk_et_al_Science_2022.pdf"
    },
    {
      "title": "A draft human pangenome reference",
      "url": "https://www.nature.com/articles/s41586-023-05896-x"
    }
  ],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}