Genetic modification is the direct alteration of an organism’s DNA using laboratory techniques. Unlike traditional breeding, which shuffles tens of thousands of genes between two parent organisms over many generations, genetic modification allows scientists to change, add, or delete specific genes with precision. The result is an organism with a trait that would be extremely unlikely to occur naturally or could never be achieved through conventional methods.
How It Differs From Traditional Breeding
Humans have been selectively breeding plants and animals for thousands of years, choosing individuals with desirable traits and crossing them over successive generations. This process works, but it’s slow and imprecise. Corn, for example, has a minimum generation time of 60 days under perfect conditions, and each cross shuffles all 32,000 of its genes at once. That means breeders often get unwanted traits alongside the ones they’re selecting for. And if no individual in the population naturally carries a gene for the desired trait, selective breeding simply can’t produce it.
Genetic modification bypasses both of those limitations. Scientists can modify a single gene rather than mixing entire genomes, which means far fewer unpredictable side effects. They can also pull genes from other species or synthesize them in a lab, so they aren’t limited to the genetic variation that already exists within a crop or its close relatives. A plant engineered to carry a gene from a bacterium, for instance, is called a transgenic organism. When the inserted gene comes from the same species or a closely related one, it’s called cisgenic.
The Basic Process
At its core, genetic modification involves cutting DNA at a specific location, then either removing a segment, replacing it, or inserting new genetic material. Early techniques relied on enzymes that could snip DNA and molecular tools that could splice segments together, allowing scientists to synthesize, sequence, and rearrange genetic code at the level of individual base pairs.
The first successful experiment in genetic engineering came in 1973, when biochemists Herbert Boyer and Stanley Cohen inserted DNA from one bacterium into another. That breakthrough opened the door to a wave of practical applications. By 1982, the FDA had approved the first consumer product made through genetic engineering: synthetic human insulin for diabetes. In 1994, the first genetically engineered food, a tomato designed to ripen more slowly, reached grocery shelves.
How CRISPR Changed the Field
The tool that transformed genetic modification from a specialized laboratory procedure into a widely accessible technology is CRISPR-Cas9. It works in three steps: recognition, cleavage, and repair. First, a short piece of guide RNA is designed to match a specific DNA sequence in the target gene. That guide RNA leads the Cas9 protein (essentially molecular scissors) to the exact right spot on the genome. Cas9 then cuts both strands of the DNA. Once the cut is made, the cell’s own repair machinery kicks in, either rejoining the broken ends (which typically disables the gene) or incorporating a new DNA template provided by researchers.
CRISPR is fast, relatively inexpensive, and far more precise than older methods. But it still has limitations. It relies on cutting both strands of DNA, which can occasionally introduce unintended changes at other locations in the genome.
Newer Precision Tools
Two newer approaches address those limitations. Base editing can swap one individual DNA letter for another without cutting both strands, which is important because most agriculturally valuable and disease-causing genetic variations involve a change in just a single letter of the genetic code. Prime editing goes further still. It uses a modified version of the Cas9 protein fused with a reverse transcriptase enzyme and a specially designed guide RNA that both identifies the target and carries the template for the desired edit. Prime editing can make all possible single-letter swaps, small insertions, small deletions, and combinations of these, all without breaking both DNA strands. It also produces far fewer unintended edits at off-target sites, because it requires multiple layers of matching between the guide RNA and the target DNA before any change is made.
Applications in Agriculture
The two most common engineered traits in crops are insect resistance and herbicide tolerance. Insect-resistant crops carry a gene that produces a protein toxic to specific pests, reducing the need for chemical pesticides. Herbicide-tolerant crops can survive application of a particular weed killer, making it easier for farmers to control weeds without damaging the crop itself.
A large meta-analysis published in PLOS One found that GM crops increase yields by an average of 21%. Insect-resistant varieties showed the bigger gains, boosting yields by about 25% compared to roughly 9% for herbicide-tolerant varieties. Insect-resistant crops also delivered larger reductions in pesticide use. The most widely grown GM crops globally are soybean, maize, and cotton.
Applications in Medicine
Genetic modification in medicine takes two broad forms. The first, and the one with the longest track record, involves engineering microorganisms to produce human proteins. Synthetic insulin, the very first consumer GMO product, is still manufactured this way.
The second form involves modifying a patient’s own cells or delivering corrected genes directly into the body. As of late 2024, the FDA had approved 20 gene therapy products in the United States. These therapies treat conditions ranging from inherited blood disorders and certain cancers to a rare form of inherited blindness and spinal muscular atrophy in infants. One notable approval, Casgevy, uses CRISPR-based editing to treat sickle cell disease, marking the first time a CRISPR therapy received regulatory clearance.
Somatic vs. Germline Editing
When gene therapy modifies a patient’s body cells (somatic cells), those changes affect only that individual. The edits are not passed on to children. This is the basis for all currently approved gene therapies.
Germline editing, by contrast, alters DNA in eggs, sperm, or embryos, meaning any changes would be inherited by future generations. That distinction raises significant ethical concerns. If an error occurs during germline editing, the consequences wouldn’t be confined to one person but could ripple through a family line indefinitely. For this reason, germline editing in humans remains broadly prohibited or heavily restricted in most countries, even as somatic gene therapies continue to advance.
Labeling and Regulation
More than 60 countries require some form of labeling or disclosure for genetically modified foods. In the United States, the National Bioengineered Food Disclosure Standard took full effect in January 2022. Under this standard, a food is considered bioengineered if it contains genetic material modified through laboratory DNA techniques in ways that could not be achieved through conventional breeding or found in nature.
There are several nuances worth knowing. Foods containing less than 5% bioengineered material are exempt from labeling. So are foods served in restaurants, cafeterias, and on airplanes. Small manufacturers with less than $2.5 million in annual sales are also exempt. And here’s a detail that surprises many people: if a food is derived from a bioengineered crop but no modified DNA remains detectable in the final product (as with highly refined oils or sugars), it does not require disclosure.
Manufacturers who do need to label their products can choose from four disclosure methods: a text statement on the package, a QR code with an accompanying phone number, a text-message option, or a standardized symbol. For very small packages, alternative options like a phone number or website are permitted.