Thirty megabases of DNA sequence from the human X chromosome will be generated over three years at a cost of less than 35 cents per finished base. The data will be of high quality, with less than one error per 10,000 nucleotides. High resolution cosmid maps from the Xq28 and Xp22 regions will be generated by automated restriction endonuclease digestion mapping. Selected cosmids will be analyzed by a balanced random and directed strategy based upon double ended sequencing of subclones that will minimize redundancy and dependence on expensive oligonucleotide walks, and avoid excessive redundant sequencing in the overlapping regions. Gaps and ambiguities in the cosmid maps will be solved by isolating alternative genomic clones, and by coordinating ongoing partial sequencing with the mapping. New technologies to improve large scale sequencing will be developed. Simplified construction of high quality shotgun libraries and a rapid DNA template preparation scheme will allow partial automation of front end steps. Novel fluorescent dye labeled oligonucleotide primers will enable streamlined sequence reactions and the resulting signal enhancements will simplify DNA base calling. The informatics infrastructure for the coordination of all steps from the cosmid mapping, DNA sequencing, data assembly, validation, annotation and database submission will be streamlined to allow higher throughput. In addition to the DNA sequence, this project will generate a model for an expanded program that can allow timely sequencing of the whole human genome.
