An approach to global restriction mapping is described that is applicable to any complex source DNA. By analyzing a single restriction digest for each member of a redundant set of A clones, a data base is constructed that contains fragment-size lists for all the clones. The clones are then grouped into subsets, each member of which is related to at least one other member by a significant overlap. Finally, a tree-searching algorithm seeks restriction maps that are consistent with the fragment-size lists for all the clones in each subset. The feasibility of the approach has been demonstrated by collecting data on 5000 A clones containing random 15-kilobase inserts of yeast DNA. It is shown that these data can be analyzed to produce regional maps of the yeast genome, extending in some cases for over 100 kilobases. In combination with hybridization probes to previously cloned genes, these local maps are already useful for defining the physical arrangement of closely linked genes. They may in the future serve as building blocks for the construction of a continuous global map.Fifteen years after the discovery of site-specific restriction endonucleases, the restriction map has become the standard medium for displaying the functional organization of short segments of DNA. For regions spanning up to 50 kilobases (kb), restriction maps are routinely constructed and have become indispensable tools both for organizing existing data and for planning experiments. In a few favorable cases, maps as long as 600 kb have been constructed at a resolution of a few kb (1-3), but it has proven difficult to extend existing mapping methods beyond this range. Given that the average human chromosome contains 105 kb of DNA, there is a gross disparity between the present limits of restriction-mapping techniques and the sequence complexity of cellular genomes.In this paper, we describe a project that is aimed at constructing a restriction map of total nuclear DNA from the yeast Saccharomyces, whose haploid genome size has been estimated to be 1.5 x 104 kb (4). The data collection has involved picking X clones at random and measuring the sizes of the restriction fragments generated by a single digest of each clone. Because enough clones have been analyzed to provide redundant sampling of most regions of the genome, it is possible to extract mapping information from the unordered fragment-size lists by an algorithm that systematically imposes the requirement that the set of restriction fragments generated from a single clone must be contiguous.At the present stage of the yeast project, we have analyzed 5000 X clones with average insert sizes of 15 kb, thereby achieving a sampling redundancy on the order of five. Using the data, it is now possible to construct maps of from 20 to >100 kb around most arbitrary starting points in the yeast genome. These local maps have already defined the physical arrangement of clusters of genes that previously had only been linked genetically, and they point the way toward the development of a global phy...

No abstract

A new method for preparing small quantities of lambda DNA from phage lysates has been developed. The protocol is based on the concentration and purification of bacteriophage particles from crude lysates using small DEAE-cellulose columns. This chromatographic step gives an absolute separation of the lambda DNA from the cellular nucleic acids and a 20-fold enrichment relative to the major soluble proteins in crude lysates, while effecting a 10-fold concentration of the phage. Final deproteinization and concentration of the lambda DNA is achieved by conventional precipitation steps. The lambda DNA produced by this method is shown to be nondegraded, biologically active, and an excellent substrate for restriction enzymes. A detailed protocol is provided for starting with individual plaques and using the method to obtain purified DNA from large numbers of lambda clones.

Physical maps of the six smallest chromosomes of Saccharomyces cerevisiae are presented. In order of increasing size, they are chromosomes I, VI, III, IX, V and VIII, comprising 2.49 megabase pairs of DNA. The maps are based on the analysis of an overlapping set of lambda and cosmid clones. Overlaps between adjacent clones were recognized by shared restriction fragments produced by the combined action of EcoRI and HindIII. The average spacing between mapped cleavage sites is 2.6 kb. Five of the six chromosomes were mapped from end to end without discontinuities; a single internal gap remains in the map of chromosome IX. The reported maps span an estimated 97% of the DNA on the six chromosomes; nearly all the missing segments are telomeric. The maps are fully cross-correlated with the previously published SfiI/NotI map of the yeast genome by A. J. Link and M. V. Olson. They have also been cross-correlated with the yeast genetic map at 51 loci.