Short-Insert Libraries as a Method of Problem Solving in Genome Sequencing

As the Human Genome Project moves into its sequencing phase, a serious problem has arisen. The same problem has been increasingly vexing in the closing phase of the Caenorhabditis elegans project. The difficulty lies in sequencing efficiently through certain regions in which the templates (DNA substrates for the sequencing process) form complex folded secondary structures that are inaccessible to the enzymes. The solution, however, is simply to break them up. Specifically, the offending fragments are sonicated heavily and recloned, as much smaller fragments, into pUC vector. The sequences obtained from the resulting library can subsequently be assembled, free from the effects of secondary structure, to produce high-quality, complete sequence. Because of the success and simplicity of this procedure, we have begun to use it for the sequencing of all regions in which standard primer walking has been at all difficult. [The sequence data described in this paper have been submitted to the EMBL data library under accession nos. Z93392 , Z92540 , and Z81558 .]