DNA, or deoxyribonucleic acid, is the molecule in living things that contains the coding information for creating proteins. It is also the molecule of heredity--whenever an organism reproduces, each offspring gets a copy of its parents' DNA.
Structurally, nucleic acids are giant polymers composed of units called nucleotides, each of which consists of a sugar, a phosphate, and a base. The sugar in DNA is deoxyribose. The four bases are adenine, cytosine, thymine, and guanine. Each can readily form hydrogen bonds to one of the others--A with T, C with G--so that two matching strands of DNA will group together in a double helix. This form is called B-DNA; there are a few others.
Reasons why thymine is used in DNA, and uracil in RNA need some explanation. Thymine is created from uracil in a methylation process, which is energetically expensive. In other words, uracil is cheaper for a cell that thymine. Unfortunatelly cytosine becomes deaminated with significant speed and becomes uracil. If uracil were used in DNA, that would be source of many mutations. With thymine, all uracils created in such way can be removed, and the proper fragment of DNA can be copied from the other strand. Cheaper uracil is used in RNA because RNA doesn't need to last a long time anyway.
DNA does not code for any proteins directly. Instead, matching strands of messenger RNA (mRNA) are formed by enzymes called RNA polymerases--this is called "transcription"--which are sent off into the cytoplasm for translation in the ribosomes. Other kinds of RNA are also formed to help in the process--tRNA, rRNA, and snRNA. The genetic code used by these molecules to encode the sequence of proteins is highly conserved, nearly universally so. This suggests that all life on Earth is descended from a single original strain of living matter.
In bacteria there is a single main strand of DNA, copied over when the cell replicates. There may also be small pieces of circular DNA called plasmids floating in the cytoplasm that can be exchanged between cells. In eukaryotes there are typically multiple pieces of DNA contained in a central nucleus, separated out in cell division by a process called mitosis. There may be a single copy of the genome (whence the organism is haploid), two copies (diploid), or many copies (polyploid).
In preparation for sexual reproduction, haploid cells are produced from a diploid precursor in a process called meiosis. Eventually some haploid cells unite together to form a new diploid cell (syngamy) with half its genome from each diploid ancestor. When there are different versions of the same gene, their effects may be blended or a single version may be dominant.
The discovery of DNA
The structure of nucleic acids was discovered in the C19. Later it was found that the sugar in nucleic acid could be either ribose (RNA) or deoxyribose (DNA). In 1943, Oswald Avery proved that DNA carries genetic information and suggested DNA might actually be the gene.
In 1948, Linus Pauling discovered that many proteins take the shape of an alpha helix. In 1950, biochemist Erwin Chargaff found that the arrangement of bases in DNA varied, but the amount of certain bases always occured in a one-to-one ratio.
In England two seperate investigations into DNA took place in the 1950s. At Cambridge University, Francis Crick and James Watson were making physical models to eventually create an accurate picture of the molecule. At Kings College in London, Maurice Wilkins and Rosalind Franklin were examing X-ray diffraction images of DNA.
In 1951, Watson attended a lecture by Franklin on her work to date. She had found that DNA can exist in two forms. From this she had deduced that the phosphate part of the molecule was on the outside. Based on a imperfect recollection of her information, Watson and Crick made a failed model. Franklin had also found that the X-ray diffractions of one of the forms of DNA had the characteristics of a helix. She suspected that DNA was helical but did not announce this.
Watson and Crick made a crucial step, suggesting the molecule was made of two nucleotide chains, each in a helix as Franklin had found, but one going up and the other going down. Crick learned of the work of Chargaff in 1952. He added that to the model, so that matching base pairs interlocked to keep the distance between the chains constant. This model showed that each strand of the DNA molecule was a template for the other. The fit with experimental data was such that it was almost immediately accepted. Their paper 'A structure for Deoxyribose Nucleic Acid' was published in April 1953. further research uncovered how DNa made proteins in 1957.