With respect to cloning of DNA, specific sequences may also be selected in some cases based on expression patterns, detectable, for example with antibodies to a specific protein.
The principal categories of genetic manipulation are producing useful products, and manipulating plants and animals. Examples of useful products include clotting factor, human insulin, human growth hormone. Producing such products often involves cooperation between biotechnology companies and pharmaceutical companies, which have experience in human testing and approvals. Useful products can be produced from bacteria using cDNAs as inserts. Genomic DNA does not work because introns cannot be processed in bacteria.. Yeast has some advantages for producing useful products, including well-developed technology for growing cells (e.g., from brewing industry) and the ability to make some protein modifications which bacteria cannot.
The genetic manipulation of plants and animals involves introducing foreign DNA into the germline of an organism, breeding the resulting individual, and producing a "transgenic" line with the foreign DNA hopefully being expressed. Vectors derived from the host organism are often used for introducing the gene. Examples include P element in Drosophila, Ti plasmid in plants, and retroviruses in mammals. Manipulation of embryonic stem cells can be used to produce "knockout" lines of mice which have specific genes inactivated. This involves homologous recombination of an inactive form of the gene in the stem cells to replace the normal form, and then introduction of the stem cells into mice to develop a knockout individual which can be bred to develop a kockout line. Commercial applications of transgenics include plants resistant to herbicides, and fast-growing strains of fish with extra growth hormone gene copies. There is controversy about the application of these methods commercially to plants and animals.
Genome analysis involves analyzing the structure of chromosomes at the DNA sequence level. The ultimate goal is to identify and analyze the gnome of an organism. Artificial chromosome vectors such as BACs are typically used in this procedure. The inserts are fingerprinted to develop markers (e.g., using sequence tagged sites, STSs or by digestion with restriction enzymes) along the length of the inserts. By using overlapping inserts (for example, by producing the library by partial digestion with restriction enzyme) it is possible to order the inserts into a "contigs", ordered sets of adjacent inserts.
Sequence analysis is in the domain of the field of "bioinformatics". One level of analysis is to look for "open reading frames" or ORFs which lack a stop codon and thus could code for a protein. A second level of analysis is to compare sequences to known sequences in databases such as Genbank of the US National Institutes of Health.