A genome is the complete set of genetic instructions an organism carries, encoded in DNA (or, for some viruses, RNA). In humans, that means roughly 3.05 billion base pairs of DNA spread across 23 pairs of chromosomes, plus a small loop of DNA inside each mitochondrion. But “complete set of genetic instructions” is deceptively simple phrasing for something biologists have spent decades redefining as they discover just how much complexity hides inside those base pairs, and how slippery the boundaries of “complete” turn out to be.
What a Genome Actually Contains
When people hear “genome,” they often picture genes lined up like beads on a string. In reality, protein-coding genes make up a small fraction of most genomes. In humans, about 20,000 protein-coding genes account for roughly 1.5 percent of total DNA. The rest used to be dismissed as “junk DNA,” a label that has aged poorly. Large stretches consist of repetitive sequences, many of which are transposable elements, sometimes called jumping genes. These mobile DNA segments can copy themselves and insert into new locations, and over evolutionary time they have reshaped genomes dramatically. The variation in genome size across species correlates more closely with the amount of repetitive DNA than with the number of protein-coding genes.1PubMed. Transposable elements and the evolution of genome size in eukaryotes Transposable elements can cause mutations, alter how genes are expressed, trigger chromosome rearrangements, and inflate genome size as their copy numbers climb.2Communications Biology. Repetitive DNA sequence detection and its role in the human genome
Beyond repetitive elements, the genome contains regulatory sequences, promoters, enhancers, and long stretches of non-coding RNA. Long non-coding RNAs, for instance, do not produce proteins but play roles in organizing chromatin, regulating when and where genes turn on, and controlling what happens to RNA molecules after they are made.3PubMed Central. The role of long non-coding RNAs in genome formatting and expression These elements blur the line between “gene” and “not a gene,” which is one reason estimates of how much of the genome is functional vary so widely.
How Much of the Human Genome Is Functional
This question has generated real scientific disagreement. One approach looks at evolutionary conservation: DNA sequences that have been preserved over millions of years of evolution are probably doing something important, because natural selection would have let useless sequences drift and change. Using DNA shape and topography rather than just raw sequence comparisons, one study estimated that about 12 percent of human DNA bases are under evolutionary constraint, roughly double what sequence-based methods alone detect.4PubMed Central. Local DNA topography correlates with functional noncoding regions of the human genome A different statistical approach that integrated multiple types of annotation data predicted about a third of the genome as functional.5Scientific Reports. A Statistical Framework to Predict Functional Non-Coding Regions in the Human Genome Through Integrated Analysis of Annotation Data
The gap between 12 percent and 33 percent is not a sign that someone is wrong. It reflects different definitions of “functional.” If you mean sequences under strong enough evolutionary pressure that their loss would be harmful, you get a smaller number. If you include sequences that are biochemically active in some context, even transiently, the estimate balloons. The ENCODE project famously reported that up to 80 percent of the genome has some biochemical activity, a figure that drew sharp criticism from researchers who argued that random biochemical noise should not count as function. The honest answer is that nobody has settled this debate, and the percentage you get depends entirely on where you draw the line between “doing something useful” and “doing something detectable.”
Genomes Beyond the Nucleus
The nuclear genome gets most of the attention, but it is not the only genome in your cells. Mitochondria, the organelles that generate most of your cellular energy, carry their own small circular DNA molecules. In humans, the mitochondrial genome is about 16,500 base pairs and contains 37 genes. Plants have an extra layer: their chloroplasts, the organelles responsible for photosynthesis, also carry their own genomes. Both chloroplasts and mitochondria replicate their DNA independently and pass it on to daughter organelles, producing a pattern of inheritance that operates outside the usual chromosomal rules.6PubMed Central. Why chloroplasts and mitochondria retain their own genomes and genetic systems: Colocation for redox regulation of gene expression
These organelle genomes are remnants of an ancient merger. Mitochondria descended from a free-living bacterium that was engulfed by an ancestral cell roughly two billion years ago, and chloroplasts trace back to a cyanobacterium captured in a similar event. Over time, most of the genes from those original bacterial genomes migrated into the host cell’s nuclear chromosomes, but a handful stayed behind. One hypothesis for why some genes remain in the organelle is that having certain genes physically close to the energy-producing machinery allows rapid, local adjustments to gene expression in response to changing energy demands.
Gene transfer between organelle genomes and the nucleus has not stopped entirely. Studies of plant species have found homologous DNA fragments shared between chloroplast and mitochondrial genomes, evidence of ongoing intracellular gene transfer.7PubMed Central. Interactions between chloroplast and mitochondrial genomes in 11 Salix species So when we talk about “the genome” of a plant or animal, we should really mean all of these DNA molecules taken together, not just the chromosomal set in the nucleus.
The Completed Human Genome
The Human Genome Project, declared finished in 2003, actually left about 8 percent of the human genome unread. Those missing regions were mostly highly repetitive stretches, including centromeres, the structural hubs of chromosomes, and the short arms of certain chromosomes. These areas resisted the sequencing technology of the time. It was not until 2022 that the Telomere-to-Telomere (T2T) Consortium published a truly complete, gapless human genome sequence: 3.055 billion base pairs, with no gaps remaining on any autosome. That effort added nearly 200 million base pairs of new sequence and identified 1,956 predicted genes, 99 of which appear to code for proteins.8PubMed Central. The complete sequence of a human genome
Filling in those final 8 percent mattered because the newly sequenced regions include some of the most structurally important parts of chromosomes. Centromeres, for instance, are critical for cell division, and their satellite repeat arrays had been invisible to earlier projects. Having the full picture opened up research into regions that had simply been off-limits before.
Why One Genome Is Not Enough
A single reference genome, no matter how complete, cannot capture the full genetic diversity of a species. The human reference genome is stitched together mostly from one person’s DNA, which means it reflects one particular set of variants. Across a species, individuals carry different alleles, structural rearrangements, and sometimes entire genes that are present in some members but absent in others. The concept of a pangenome addresses this limitation.
A pangenome is the full set of all genes present across a species or defined group of organisms. It breaks down into a core genome, the genes shared by every individual, and an accessory genome, the genes found in only some individuals.9PubMed. The Ecology and Evolution of Pangenomes The idea originated in microbiology, where bacteria of the same species can differ wildly in gene content. A single species of bacterium might have a core genome of 1,000 genes shared universally, plus thousands of accessory genes distributed patchily across different strains.10PubMed Central. First Steps in the Analysis of Prokaryotic Pan-Genomes
The pangenome concept has since expanded to plants, animals, and humans. In 2023, a human pangenome reference was released, built from 47 diverse individuals. This matters for medicine because a single reference genome can introduce biases: if your DNA differs from the reference in ways that are normal for your population, variant-calling software might flag those differences as mutations when they are simply part of human diversity. A pangenome reference makes genetic analysis more equitable and more accurate for people of all backgrounds.11Briefings in Bioinformatics. A gentle introduction to pangenomics
The Genome Size Puzzle
If genomes were efficiently organized blueprints, you would expect more complex organisms to have bigger genomes. They do not. This mismatch, sometimes called the C-value paradox, refers to the lack of any tidy correlation between an organism’s biological complexity and the amount of DNA it carries.12PubMed Central. C-value paradox: Genesis in misconception that natural selection follows anthropocentric parameters of ‘economy’ and ‘optimum’ A single-celled amoeba can have a genome hundreds of times larger than a human’s. Some ferns have genomes ten times our size. The marbled lungfish holds the record among vertebrates at over 130 billion base pairs.
Among eukaryotes, genome size varies more than 60,000-fold, and that variation is not explained by differences in gene number.13PubMed Central. What’s in a genome? The C-value enigma and the evolution of eukaryotic genome content Most of the difference comes from non-coding sequences, especially transposable elements. Organisms with small genomes tend to keep transposable element copy numbers low and tightly controlled, while organisms with large genomes have relaxed those constraints, allowing mobile elements to proliferate.14PubMed. Transposable elements and the evolution of genome size in eukaryotes Whether this represents an evolutionary advantage, a tolerated burden, or something in between remains debated.
Your Genome Is Not Identical in Every Cell
The textbook picture says that every cell in your body carries the same genome, copied faithfully from the original fertilized egg. The reality is messier. Every time a cell divides, it has a chance of introducing new mutations. Over a lifetime, trillions of cell divisions accumulate changes, meaning that by adulthood your body is a patchwork of slightly different genomes. This is called somatic mosaicism, and it is universal: every individual is genetically different from cell to cell.15PubMed Central. Somatic Mosaicism in Biology and Disease
Most somatic mutations are harmless. Some, though, land in genes that control cell growth, which is how many cancers begin. Others accumulate in brain cells and may contribute to neurological conditions. Advances in sequencing now allow researchers to detect these tiny differences between cells, revealing that the genome is less a fixed blueprint and more a living document that accumulates edits over a lifetime.
Information Layered on Top of the Sequence
Even a complete, perfectly sequenced genome does not tell you everything about how an organism works. Chemical modifications attached to DNA and to the histone proteins that DNA wraps around can change how genes behave without altering the underlying sequence. These modifications, things like methyl groups added to specific DNA bases or acetyl groups tagged onto histones, influence whether a gene is accessible and active or tightly packed and silent. The study of these modifications across the whole genome is called epigenomics, and it encompasses features like DNA methylation patterns and histone modification landscapes.16PubMed Central. The distinction between epigenetics and epigenomics
Epigenetic information interacts with chromatin (the DNA-protein complex) without changing the DNA sequence itself, yet it profoundly affects gene function.17PubMed Central. The impact of epigenetic information on genome evolution This is why a liver cell and a neuron can contain identical DNA but behave completely differently: their epigenomic landscapes differ. It is also why environmental factors like diet, stress, and chemical exposures can alter gene activity without mutating DNA. When someone asks “what is a genome,” the practical answer increasingly has to include these layers of information sitting on top of the raw sequence.
Three-Dimensional Genome Organization
DNA is not a flat string stretched out inside the nucleus. It is folded into an elaborate three-dimensional structure, and that folding matters. Techniques that capture how chromosomes physically interact with each other have revealed that genomes are organized into regions called topologically associating domains, or TADs. These are stretches of DNA that interact with themselves much more frequently than with neighboring regions, forming a kind of structural neighborhood.18PubMed Central. Principles of genome folding into topologically associating domains
TADs are not just architectural curiosities. They play a role in gene regulation by keeping enhancers (sequences that boost gene activity) in close physical contact with the genes they are meant to control, while insulating them from genes in neighboring domains. Disrupting TAD boundaries, through deletions or rearrangements, can cause enhancers to activate the wrong genes, which has been linked to developmental disorders and cancers.19PubMed Central. Uncovering topologically associating domains from three-dimensional genome maps with TADGATE The genome’s definition, then, is not just a matter of sequence. Physical folding is itself a form of genomic information.
Minimal Genomes and the Boundary of Life
One way to understand what a genome really needs is to strip it down to the bare essentials. Researchers at the J. Craig Venter Institute did exactly that, building a synthetic bacterium with the smallest genome of any self-replicating cell found in nature: 531 kilobase pairs containing just 473 genes.20PubMed. Design and synthesis of a minimal bacterial genome They arrived at this through iterative rounds of designing, building, and testing, removing genes until the cell could barely survive and reproduce in a laboratory setting.
Even at that minimal size, roughly a third of those 473 genes had no known function at the time. The cell needed them to live, but researchers could not explain why. A separate theoretical effort estimated that a minimal genome could be as small as 113 kilobase pairs with 151 genes, though that estimate assumed an idealized environment supplying many nutrients externally.21PubMed Central. Towards synthesis of a minimal cell These experiments highlight something fundamental about the genome concept: the minimum size of a genome depends on the environment the organism inhabits. A free-living bacterium needs more genes than a parasite that steals nutrients from a host, and both need more than a virus that hijacks its host’s machinery entirely.
Genomes Without Organisms
The definition of a genome gets stretched even further in environmental genomics. Most microbes on Earth have never been cultured in a lab, which means their genomes were unknown until researchers started sequencing DNA directly from environmental samples. This approach, called metagenomics, skips the step of growing an organism and instead reconstructs genomes computationally from mixed community DNA. A major database now integrates over 2.7 million metagenome-assembled genomes from more than 100,000 samples covering ecosystems from the human gut to ocean floors to extreme environments, capturing over 109,000 species-level clusters. Roughly 63 percent of those clusters represent previously uncharacterized species.22PubMed Central. gcMeta 2025: a global repository of metagenome-assembled genomes enabling cross-ecosystem microbial discovery and function research
These metagenome-assembled genomes are not isolated from a single organism. They are computational reconstructions pieced together from short DNA fragments, and they can contain errors or chimeric sequences. Still, they have revolutionized microbiology by revealing that the known tree of life was missing most of its branches. The practical meaning of “genome” in this context is something more like “a reconstructed genetic blueprint of something that probably exists as a distinct organism in nature,” which is a far cry from the tidy definition found in a textbook.
Ancient Chromosomes and Deep Conservation
Genomes change over evolutionary time, but some features are remarkably persistent. Comparative analysis of chromosome-scale genomes across animals as different as sponges, corals, and vertebrates has revealed conservation of synteny, meaning that large blocks of genes remain on the same chromosome, in roughly the same arrangement, across hundreds of millions of years of evolution.23PubMed Central. Deeply conserved synteny and the evolution of metazoan chromosomes This allows researchers to reconstruct the chromosomes of ancient ancestors that left no fossils, building a picture of what the genome of the last common ancestor of all animals might have looked like.
This deep conservation suggests that chromosome organization itself is under some form of evolutionary constraint, not just the individual genes. Rearranging which genes sit near each other on a chromosome can disrupt regulatory relationships, which may be why certain gene neighborhoods have persisted across vast stretches of evolutionary time. The genome, viewed this way, is not just a parts list. The arrangement of those parts carries biological meaning that evolution has preserved.
Sequencing Technology and What Counts as “Known”
How we define a genome is inseparable from how we read one. Early sequencing methods produced short reads, small fragments of a few hundred bases that had to be assembled like a puzzle. These approaches worked well for unique sequences but failed in repetitive regions, which is why the human genome had gaps for two decades. Long-read sequencing technologies changed the game by producing reads of tens of thousands of bases or more, long enough to span entire repetitive regions in a single piece. This leap in read length is what made the complete T2T human genome possible, and it has allowed researchers to assemble genomes for thousands of other species that were previously intractable.
The technology matters because our understanding of genome content has always been limited by our ability to read it. Regions that could not be sequenced were treated as terra incognita, blank spots on the map. As sequencing improves, the definition of any given organism’s genome gets revised, sometimes substantially. What counts as “the genome” at any moment is partly a statement about biology and partly a statement about current technological capability.