The Cell, the Nucleus and the Mitochondrion as DNA Reservoirs

The nucleus is bounded by a double membrane (the nuclear envelope) perforated by nuclear pore complexes through which RNA and proteins transit. Inside, the genome is organised into chromosomes that are visible as condensed structures only during cell division. In interphase, the working state of most cells, chromosomes occupy distinct territories within the nucleus and exist in a partially decondensed form that allows gene transcription.

Human somatic cells are diploid: they carry 46 chromosomes in 23 pairs. Chromosomes 1 through 22 are the autosomes, present in two copies in both males and females. Chromosome pair 23 is the sex chromosome pair: females carry two X chromosomes (46,XX), and males carry one X and one Y chromosome (46,XY). Germ cells (eggs and sperm) are haploid, carrying one copy of each chromosome, produced by meiosis. When a sperm fertilises an egg, the diploid complement is restored.

The forensic implications of this architecture are concrete. Autosomal STR typing, the method behind virtually every modern DNA database, targets repeat sequences on the autosomes. Because autosomes are inherited in pairs from both parents, an autosomal STR profile is unique to an individual (excepting identical twins and somatic mutations) and carries exceptional discriminating power. The random-match probability for a full CODIS 20-locus profile in an unrelated individual is typically one in many quadrillions. In the US, the NDIS (National DNA Index System) holds over 21 million offender and arrestee profiles, all based on autosomal STR loci. In the UK, the National DNA Database (NDNAD) holds profiles typed at 16 ESS-compatible loci. India's proposed national database under the DNA Technology (Use and Application) Regulation Bill 2019 would use a set of loci compatible with international exchange protocols. The marker classes that underpin these databases and their inheritance rules are covered in polymorphism, heredity and forensic markers. The biochemical properties of the double helix that make PCR recovery possible are described in DNA, RNA and the central dogma.

The Y chromosome is paternally inherited as a non-recombining haplotype (with the exception of the pseudoautosomal regions at the chromosome tips). All males in a direct paternal lineage share the same Y-chromosome haplotype unless a mutation has occurred. This makes Y-STR profiling valuable for male-lineage identification and for separating a male contributor from a female-dominated mixture, but it means a Y-STR result cannot distinguish an alleged perpetrator from his father, brother, or paternal uncle, the statistics and casework for Y-STR and X-STR profiling are covered in Module 6. The YHRD (Y-chromosome Haplotype Reference Database) holds over 300,000 haplotypes across global populations and provides the statistical framework for reporting Y-STR results.

Mitochondria are cytoplasmic organelles, roughly 1-10 micrometres in length, responsible for oxidative phosphorylation and ATP production. Each mitochondrion contains two to ten copies of its genome, and a single cell may contain hundreds to thousands of mitochondria depending on the tissue's energy demand. Oocytes are particularly mitochondria-rich, with estimates of 100,000 to 600,000 mitochondrial genomes per cell. Mature red blood cells have no mitochondria at all: they expel all organelles during differentiation, which is why blood-derived DNA is nuclear only.

The mitochondrial genome (mtDNA) is circular, double-stranded, and 16,569 base pairs in length in humans. It encodes 13 protein subunits of the oxidative phosphorylation complexes, 22 transfer RNAs, and 2 ribosomal RNAs. The non-coding control region (the D-loop, approximately base pairs 16024-576) contains the origin of replication and the promoters for transcription. Within the D-loop sit two hypervariable regions, HV1 (roughly bp 16024-16365) and HV2 (roughly bp 73-340), that accumulate substitutions at a rate approximately ten times higher than the nuclear genome average. Forensic mtDNA typing targets HV1 and HV2 sequencing and compares the resulting profile against the revised Cambridge Reference Sequence (rCRS, published 1999), reporting differences as numbered positions.

Two features of mtDNA inheritance drive its forensic utility and its limitations. First, mitochondrial genomes are inherited almost exclusively through the maternal line: the mitochondria contributed by the sperm cell are systematically destroyed in the fertilised egg by ubiquitin-mediated autophagy. All children of a mother share her mitochondrial haplotype (subject to rare mutations). A maternal lineage can therefore be traced across many generations and many missing relatives. The identification of the Romanov family remains in 1991-1994 used mtDNA sequencing targeting HV1 and HV2 to confirm that bones excavated near Yekaterinburg matched the maternal-lineage haplotype of Prince Philip, Duke of Edinburgh (a maternal-lineage relative of Tsarina Alexandra), and of two living maternal-lineage relatives of Tsar Nicholas II. The result, while not individually identifying, was combined with STR profiling to reach a court-standard conclusion.

Second, because all maternal relatives share the same mtDNA haplotype (absent mutations), an mtDNA result does not distinguish siblings, a mother and a child, or first and second maternal-lineage cousins. The EMPOP (European mtDNA Population Database) provides the statistical framework for interpreting a match as a match probability across relevant population groups, but the numbers are never as powerful as a full autosomal STR profile. In the UK, the Forensic Science Regulator's guidance requires that mtDNA results be reported with a population-based match probability drawn from a validated database. In the US, SWGDAM's mtDNA Interpretation Guidelines specify the same.

Every sample type a forensic examiner receives has a characteristic DNA content and condition that flows from the cell biology of that tissue. Understanding the tissue-level biology is what allows an examiner to choose the right extraction and amplification strategy before opening the tube.

Blood contains nucleated cells (white blood cells, or leucocytes) and non-nucleated red blood cells (erythrocytes) in roughly 1:1000 ratio by count. The DNA in a bloodstain comes entirely from the leucocytes. A single microlitre of whole blood yields approximately 25-50 nanograms of genomic DNA from roughly 4,000-8,000 leucocytes. A Bloodstain from a fresh 5-cm-diameter pool may yield micrograms of DNA. Even a dried bloodstain that has been on a surface for weeks, under typical indoor conditions, often contains sufficient high-quality DNA for full-profile STR typing.

Semen from a fertile male individual contains both spermatozoa (approximately 200-600 million per millilitre of ejaculate) and non-sperm cells (epithelial cells of the male urethra, leucocytes). Sperm heads are haploid, carrying 23 chromosomes. Each sperm head contains approximately 3 picograms of DNA. Importantly, sperm cells have unusually compact chromatin: the histones of somatic cells are replaced with protamines during spermatogenesis, producing a chromatin structure that is roughly six times more condensed than somatic nuclei and significantly more resistant to enzymatic digestion and heat denaturation. This resistance is exploited by differential extraction, which selectively lyses the fragile epithelial cells first (yielding the female fraction) and then lyses the protamine-compacted sperm under harsher conditions (yielding the male fraction). Differential extraction is the standard protocol in sexual-assault casework in the US (SWGDAM guidelines), the UK (FSR guidelines), and Australian and Indian forensic DNA laboratories, the mechanics are detailed in differential extraction in sexual assault casework.

Saliva deposits on envelopes, cigarette ends, bite marks, and skin contain buccal epithelial cells shed from the oral mucosa. Each epithelial cell is diploid and contains the standard nuclear complement. Saliva itself is not enriched for DNA: the liquid component contributes nothing, and the cell count per millilitre of saliva is highly variable. Swabbing the inside of a cheek yields a reference sample of very high DNA quality. Swabbing a bitten surface may yield trace-level DNA from a small number of shed cells.

Hair shafts without roots contain no nucleated cells: the follicle is the only part of a growing hair that contains nuclear DNA (in the root cells). The hair shaft itself contains only mitochondria-rich cells that have died and keratinised, contributing only mtDNA. A shed (telogen) hair, which has lost its root, yields only mtDNA from the shaft. An anagen (actively growing) hair plucked with the root attached may yield sufficient nuclear DNA from the follicle cells for STR profiling. This distinction, lost on many investigators at the scene, determines the entire downstream analytical pathway.

Skeletal remains from burials, fires, or water submersion represent the most challenging source material. Bone cells (osteocytes, osteoblasts) carry nuclear DNA, but diagenetic degradation, mineral replacement, and microbial colonisation destroy much of it. The densest bone, the petrous portion of the temporal bone, has the highest DNA preservation because of its mineralisation density. A 2015 study by Pinhasi and colleagues demonstrated that the petrous bone was the best source of ancient DNA from archaeological specimens across multiple burial environments, a finding now standard in forensic DVI practice. In India's tropical climate, bones recovered from outdoor graves degrade more rapidly than in temperate European or North American conditions, which affects both the quantity and quality of recoverable DNA.

The standard forensic expectation is that all maternal-lineage relatives share an identical mtDNA haplotype. In practice, two complications arise. The first is heteroplasmy: the presence of more than one mtDNA sequence variant within a single individual. Because each cell carries hundreds to thousands of mtDNA copies, it is possible for some copies to carry one sequence and others to carry an alternative (usually differing by one nucleotide), a state produced by somatic mutation or by incomplete segregation during the replication of the mtDNA pool. Heteroplasmy is classified as length heteroplasmy (insertions or deletions in homopolymeric runs, common in the C-stretch of HV1) or point heteroplasmy (a mixture of two bases at a single position). Both types are observed in forensic casework and must be reported according to the examiner's guidelines: SWGDAM's mtDNA Interpretation Guidelines specify the reporting conventions for heteroplasmic positions in US casework. The EMPOP database records heteroplasmic positions and incorporates them into match definitions.

The second complication is paternal leakage. While ubiquitin-mediated destruction of paternal mitochondria in the fertilised egg is the normal mammalian mechanism, rare exceptions have been documented. Schwartz and Vissing reported a patient in 2002 (New England Journal of Medicine) whose muscle cells contained a substantial proportion of paternal mtDNA, resulting from a failure of the normal elimination mechanism. Such cases are extremely rare, but they inform the interpretive framework: forensic examiners are trained to treat an apparent single-base mismatch between a questioned specimen and a reference mtDNA profile as a possible heteroplasmy or paternal leakage event rather than an automatic exclusion, pending confirmation by re-sequencing on an independent extract.

The forensic community operationalised these rules most clearly in the Daubert hearings surrounding early mtDNA casework in the US (State v. Ware, 1999; US v. Beverly, 2002). Courts across US jurisdictions have now largely accepted mtDNA evidence as scientifically valid when accompanied by a qualified statistician's interpretation using validated population databases. In the UK, forensic mtDNA analysis is conducted by the National Crime Agency's Forensic Capability Network and by accredited private providers, all under FSR oversight. In India, CFSL Hyderabad and CFSL Chandigarh have both published internal validation studies for mtDNA sequencing.