Wikipedia has some useful pages on SOLiD Sequencing and 2-Base Encoding

SOLiD (Sequencing by Oligonucleotide Ligation and Detection), as its name suggests, is an example of “sequencing by ligation”. The first genome sequenced with SOLiD technology was completed in 2005¹⁾.

The sample to be analyzed is ligated to an adaptor sequence and attached to beads; the DNA fragment on each bead is amplified by emulsion PCR, and the resulting polony beads are placed on a slide for sequencing. Ligase and a primer matching the adaptor sequence is added. A mixture of fluorescent “di-base probes” is added, and probes which match the next two bases on the growing chain are ligated. The fluor is then excited, the color of the light produced is recorded, and the fluor is cleaved from the end of the probe, allowing a new set of probes to attach. The cycles of hybridization/ligation/imaging are continued until the maximum read length is achieved (~50bp for SOLiD 4)²⁾. The strands are then denatured, a new “n-1” primer is added which is shifted by one base pair from the previous primer, and the ligation sequencing steps are repeated. The whole process is repeated with n-2, n-3, and n-4 primers. As a result of these multiple “resequencing” steps, each base pair is sequenced twice. This redundancy is necessary for converting the 4-color dye code into the 16 possible dinucleotide pairs.

Aside from the massive sequence coverage produced by SOLiD runs (up to 300M reads of 50bp in practice, for SOLiD 3+ systems)and high degree of accuracy, a major advantage of SOLiD sequencing is its native incorporation of barcoding and paired-end (“mate-pairs”) sequencing. However, the short read length is a huge disadvantage for assembling genomes or highly repetitious sequences.

- Text from a draft of Jenny Draper's Doctoral Thesis. Please change. — Jenny Draper 2010/03/30 16:30