Gene targeting is a biotechnological tool used to change the DNA sequence of an organism (hence it is a form of genome editing). It is based on the natural DNA-repair mechanism of homology directed repair (HDR), including homologous recombination. Gene targeting can be used to make a range of sizes of DNA edits, from larger DNA edits such as inserting entire new genes into an organism, through to much smaller changes to the existing DNA such as a single base-pair change. Gene targeting relies on the presence of a repair template to introduce the user-defined edits to the DNA. The user (usually a scientist) will design the repair template to contain the desired edit, flanked by DNA sequence corresponding (homologous) to the region of DNA that the user wants to edit; hence the edit is targeted to a particular genomic region. In this way Gene Targeting is distinct from natural homology-directed repair, during which the 'natural' DNA repair template of the sister chromatid is used to repair broken DNA (the sister chromatid is the second copy of the gene). The alteration of DNA sequence in an organism can be useful in both a research context – for example to understand the biological role of a gene – and in biotechnology, for example to alter the traits of an organism (e.g. to improve crop plants).

To create a gene-targeted organism, DNA must be introduced into its cells. This DNA must contain all of the parts necessary to complete the gene targeting. At a minimum this is the homology repair template, containing the desired edit flanked by regions of DNA homologous (identical in sequence to) the targeted region (these homologous regions are called "homology arms" ). Often a reporter gene and/or a selectable marker is also required, to help identify and select for cells (or "events") where GT has actually occurred. It is also common practice to increase GT rates by causing a double-strand-break (DSB) in the targeted DNA region. Hence the genes encoding for the site-specific-nuclease of interest may also be transformed along with the repair template. These genetic elements required for GT may be assembled through conventional molecular cloning in bacteria.

Gene targeting methods are established for several model organisms and may vary depending on the species used. To target genes in mice, the DNA is inserted into mouse embryonic stem cells in culture. Cells with the insertion can contribute to a mouse's tissue via embryo injection. Finally, chimeric mice where the modified cells make up the reproductive organs are bred. After this step the entire body of the mouse is based on the selected embryonic stem cell.

To target genes in moss, the DNA is incubated together with freshly isolated protoplasts and with polyethylene glycol. As mosses are haploid organisms, moss filaments (protonema) can be directly screened for the target, either by treatment with antibiotics or with PCR. Unique among plants, this procedure for reverse genetics is as efficient as in yeast. Gene targeting has been successfully applied to cattle, sheep, swine and many fungi.

The frequency of gene targeting can be significantly enhanced through the use of site-specific endonucleases such as zinc finger nucleases, engineered homing endonucleases, TALENS, or most commonly the CRISPR-Cas system. This method has been applied to species including Drosophila melanogaster, tobacco, corn, human cells, mice and rats.

The relationship between gene targeting, gene editing and genetic modification is outlined in the Venn diagram below. It displays how 'Genetic engineering' encompasses all 3 of these techniques. Genome editing is characterised by making small edits to the genome at a specific location, often following cutting of the target DNA region by a site-specific-nuclease such as CRISPR. Genetic modification usually describes the insertion of a transgene (foreign DNA, i.e. a gene from another species) into a random location within the genome. Gene-targeting is a specific biotechnological tool that can lead to small changes to the genome at a specific site - in which case the edits caused by gene-targeting would count as genome editing. However gene targeting is also capable of inserting entire genes (such as transgenes) at the target site if the transgene is incorporated into the homology repair template that is used during gene-targeting. In such cases the edits caused by gene-targeting would, in some jurisdictions, be considered as equivalent to Genetic Modification as insertion of foreign DNA has occurred.

Gene targeting is one specific form of genome editing tool. Other genome editing tools include targeted mutagenesis, base editing and prime editing, all of which create edits to the endogenous DNA (DNA already present in the organism) at a specific genomic location. This site-specific or 'targeted' nature of genome editing is typically what makes genome-editing different to traditional 'genetic modification' which inserts a transgene at a non-specific location in the organisms' genome, as well as gene-editing making small edits to the DNA already present in the organisms, verses genetic modification insertion 'foreign' DNA from another species.

Because gene editing makes smaller changes to endogenous DNA, many mutations created through genome-editing could in theory occur through natural mutagenesis or, in the context of plants, through mutation breeding which is part of conventional breeding (in contrast the insertion of a transgene to create a Genetically Modified Organism (GMO) could not occur naturally). However, there are exceptions to this general rule; as explained in the introduction, GT can introduce a range of possible size of edits to DNA; from very small edits such as changing, inserting or deleting 1 base-pair, through to inserting much longer DNA sequences, which could in theory include insertion of an entire transgene. However, in practice GT is more commonly used to insert smaller sequences. The range of edits possible through GT can make it challenging to regulate (see Regulation).