An organism's inherited DNA determines certain characteristics through directing the production of proteins and RNA molecules involved in protein synthesis. Proteins, in other words, serve as the connection between genotype and phenotype.

The mechanism through which DNA controls the production of proteins is known as gene expression (or, in some cases, just RNAs). Transcription and translation are two phases in the expression of genes that code for proteins. Genetic mutations impact animals via their proteins.

Archibald Garrod, a British physician, was the first to propose in 1902 that genes control phenotypes via enzymes, proteins that catalyze particular chemical processes in the cell. He proposed that the symptoms of an inherited disease are caused by an inability to produce a certain enzyme. He later referred to such disorders as "inborn errors of metabolism." People with alkaptonuria have black urine because it includes alkapton, a substance that darkens when exposed to air.

Garrod reasoned that most individuals have an enzyme that breaks down alkapton, but those with alkaptonuria have a hereditary inability to produce that enzyme, thus alkapton is excreted in their urine.

Several decades later, studies corroborated Garrod's idea that a gene determines the production of a single enzyme, which was eventually dubbed the one gene-one enzyme hypothesis. Biochemists discovered that most organic compounds are synthesized and degraded by cells via metabolic pathways, in which each chemical reaction in a series is mediated by a particular enzyme.

In the 1930s, George Beadle, an American scientist, and geneticist, and his French colleague Boris Ephrussi hypothesized that in Drosophila, each mutation affecting eye color inhibits pigment synthesis at a specific phase by inhibiting the development of the enzyme that catalyzes that step. However, neither the chemical processes nor the enzymes that catalyzed them were known at the time.

A breakthrough occurred a few years later at Stanford University when Beadle and Edward Tatum began studying with a haploid bread mold called Neurospora crassa.

Beadle and Tatum needed to disable just one allele (rather than two, as in a diploid animal) of a protein-coding gene necessary for a certain metabolic function to see a change in the phenotype of a mutant.

They blasted Neurospora with X-rays, which are known to trigger mutations, and searched among the survivors for mutants with nutritional requirements different from the wild-type bread mold.

Adrian Srb and Norman Horowitz, then at Stanford University, used Beadle and Tatum's experimental method to find mutants that required arginine in their growth medium when working with the mold Neurospora crassa.

The researchers discovered that these mutants were divided into three groups, each of which was deficient in a separate gene.

They suspected that the metabolic process of arginine production required a precursor nutrient and the intermediary molecules ornithine and citrulline based on investigations on mammalian liver cells, as shown in the diagram on the right.

Their most famous experiment is illustrated below, but the one gene-one enzyme theory and their proposed arginine-synthesizing route to the test.

Srb and Horowitz concluded from the mutants' growth need that each class of mutant was unable to carry out one step in the route for generating arginine, probably due to a lack of the required enzyme.

They reasoned that because each of their mutants was mutated in a single gene, each altered gene must typically regulate the synthesis of one enzyme.

Their findings corroborated the one gene-one enzyme theory presented by Beadle and Tatum, as well as confirming that the arginine route reported in mammalian livers also exists in Neurospora. (Note in the Results that a mutant may only develop if fed with a chemical produced after the faulty step)

Mutants in each class needed a distinct collection of chemicals throughout the three-step arginine-synthesis process.

These findings, along with those of many other comparable studies conducted by Beadle and Tatum, showed that each class was inhibited at a distinct point in the process because mutants in that class lacked the enzyme that catalyzed the inhibited step.

Because Beadle and Tatum designed their experiments so that each mutant was deficient in a single gene, the combined data offered strong support for a working hypothesis they had suggested previously. The one gene-one enzyme theory, as it was termed, asserts that the purpose of a gene is to direct the production of a protein.

This hypothesis received more support from experiments that determined which enzymes were missing in the mutations. In 1958, Beadle and Tatum shared the Nobel Prize in Physics for "discovering that genes work by controlling certain genes”, chemical occurrences” (in the words of the Nobel committee).

It is common practice to use linguistic words to explain the transfer of information from gene to protein.

In the same way that certain sequences of letters in a language such as English transmit information, nucleic acids and proteins are polymers with unique sequences of monomers that transfer information.

The monomers in DNA or RNA are the four kinds of nucleotides that vary in their nitrogenous bases. Genes are generally hundreds or thousands of nucleotides long, with each gene having a unique nucleotide sequence. Each polypeptide of a protein also contains monomers that are organized in a certain linear order (the protein's basic structure).