Chromosomes, Genes and the Human Genome

A chromosome contains a long DNA molecule associated with proteins and organised within the nucleus. About 147 base pairs wrap around a histone octamer in a nucleosome. Linker DNA and associated proteins connect and regulate these units. Packaging is dynamic rather than a rigid sequence of identical coils in every cell.

The familiar model showing a compulsory regular 30-nanometre fibre at every stage is an oversimplification. Studies of native chromatin reveal heterogeneous organisation, loops and domains. Mitotic chromosomes are highly compacted, while interphase chromosomes occupy territories and retain local differences in accessibility.

An X-shaped chromosome after replication contains two sister chromatids, each with its own DNA molecule, joined until segregation. Before replication, the corresponding chromosome has one chromatid. Chromosome number is conventionally counted by centromeres, so DNA amount can double without immediately doubling the chromosome count.

Sperm chromatin is largely compacted with protamines, with some histones retained. This creates extraction and preservation differences, but does not guarantee an intact profile from every sperm-containing stain. Mature red cells lack nuclei, and shed surface cells may also lack visible nuclei. A biological deposit should not be assumed to contain intact chromosomes simply because it came from human tissue.

A typical nucleated somatic cell before replication carries 22 pairs of autosomes and a sex-chromosome complement, usually giving 46 chromosomes. A normal haploid gamete carries 23. Germ cells before the relevant divisions are not all haploid, and chromosome-number variation occurs.

One haploid set contains roughly three billion base pairs. A typical diploid nucleus contains approximately twice that amount before replication. Exact totals depend on the sequence and chromosome complement being described. The mitochondrial genome is separate and has a variable copy number.

The Human Genome Project produced landmark draft and reference sequences in 2001 and 2003, but difficult regions remained unresolved. Telomere-to-telomere work subsequently filled major gaps, with a complete Y-chromosome sequence reported in 2023. Reference assemblies are versioned representations, not exact copies of every person’s genome.

There are roughly 20,000 human protein-coding genes. The protein-coding portions occupy a small fraction of the genome, around one to two percent; complete gene regions include other sequence. Saying that “all genes occupy 1.5 percent” confuses coding sequence with entire genes and non-coding RNA genes.

A gene is a genomic region that contributes to a functional product, including a protein or a functional RNA. Many protein-coding genes contain exons and introns. Exons are retained in a mature transcript and can include untranslated regions as well as protein-coding sequence.

Transcription produces a precursor RNA. Splicing removes introns in the relevant transcript, while alternative processing can produce different isoforms. The 5-prime cap and most poly-A tails are added during RNA processing; they are not simply a genomic stretch copied from a pre-existing cap to a pre-existing tail.

An intronic STR can be present in the precursor transcript even though it is removed from the mature mRNA. Intronic and intergenic variation can sometimes influence regulation, splicing or other functions. It is therefore incorrect to say that every non-coding STR has no possible effect or conveys no sensitive information.

Regulatory elements can lie near or far from the genes they influence. Centromeres support chromosome segregation and are not always at the physical centre; telomeres protect chromosome ends. Repetitive DNA includes several classes with different arrangements and biological roles.

Forensic panels select a subset of variable targets appropriate to their purpose. Limiting the targets can reduce some information exposure compared with broad sequencing, but does not remove privacy issues involving identification, kinship, database access or future associations.

Variation includes single-nucleotide changes, insertions, deletions, copy-number differences, larger structural changes and repeat-number variation. Counts depend on the individuals, reference and variant classes compared. A fixed number of SNP differences cannot describe every pair of unrelated people.

STR alleles often differ in repeat count. Replication slippage can change repeat length when strands misalign during copying, and repeat structure affects the process. A mutation in a germline lineage can be inherited, while somatic mutations may be confined to part of a person’s tissues.

Mutation rates vary by locus, allele, repeat structure and parental factors. A single universal rate such as one in 500 transmissions is not appropriate for every forensic STR. An apparent parent–child inconsistency needs assessment of mutation, null alleles, analytical error, relationships and the other results rather than automatic inclusion or exclusion from one observation.

Monozygotic twins usually share conventional inherited STR profiles, but can accumulate sequence differences through mutation. “Identical twins have exactly the same genome in every cell” is too absolute. Chimerism, mosaicism and transplantation can also complicate the assumption that every specimen from one person has an identical profile.

Population allele frequencies help assess genotype probabilities under an appropriate model. Heterozygosity describes variation; it does not by itself determine the probability of a particular complete profile. Independence assumptions, population structure and possible relatedness require explicit consideration.

Useful markers show informative population variation, reliable amplification and calling, suitable target length, characterised inheritance and manageable artefacts. Multiplex design also needs compatible primers, distinguishable products, sensitivity and interpretation rules. A locus is not suitable solely because it has many alleles.

Assess population data and appropriate equilibrium or association tests, while accounting for sample size and multiple comparisons. Loci on the same chromosome may show little population association but still require linkage-aware modelling in kinship work. Physical separation or a fixed centimorgan rule does not automatically establish every statistical assumption.

For a simple illustration only, suppose two independent loci have genotype probabilities 0.02 and 0.03 under a stated unrelated-person model. Their joint probability under that model is 0.0006. Those inputs must come from justified data; this calculation is not the probability of innocence and does not apply unchanged to mixtures or relatives.

The FBI’s US national system expanded from 13 to 20 core loci in January 2017. Other database systems and interoperability arrangements have their own target sets and eligibility rules. A panel’s compatibility with a database must be checked against that system’s actual requirements.

Harmonised markers can facilitate permitted comparisons but do not override differences in legal authority, retention, access or reporting. A forensic panel should be assessed as a whole, and new methods need appropriate validation before casework rather than a promised universal discrimination figure.

XX and XY are common human chromosome complements, but chromosome and developmental variation mean that a marker result is not a complete description of sex characteristics or gender. Amelogenin and other sex-associated targets have assay limits, including possible deletions and dropout. Report the information actually obtained.

Autosomal STRs can support donor and relationship comparisons across people. X-linked markers have inheritance and linkage patterns that can help particular kinship questions. They cannot be treated as automatically independent autosomal markers in every pedigree.

Many forensic Y markers lie in the male-specific region, where ordinary recombination is limited; pseudoautosomal regions have different behaviour. Y haplotypes can be shared within paternal lineages and among unrelated people. Mutation means that not every paternal relative must have an identical result, especially with rapidly mutating markers.

Mitochondrial sequences usually follow maternal inheritance and can also be shared by unrelated people. Heteroplasmy and mutation affect comparisons. Testing may cover specified control-region ranges or the whole mitochondrial genome; the typed range and quality need reporting.

Neither a Y nor mitochondrial agreement proves that the donor belongs to the named person’s family or identifies one member uniquely. Use appropriate reference data, lineage-aware interpretation and the relevant method guidance. Higher mitochondrial copy number may aid recovery, but it does not establish that nuclear DNA is absent from a rootless hair or degraded bone.