IN THIS ISSUE

Proteomics is a rapidly growing area of functional genomics, and there is no doubt that Craig Venter's highly publicized, if rather late, entry into this ®eld will add to the pace of technological advancement. His partners in this venture, PE Biosystems, plan to develop high speed mass spectrometers to be used to identify each individual protein from the mixtures obtained from cell or tissue samples. In this review, Paul Haynes and John Yates examine the current state of analytical methods in proteomics, in which mass spectrometry is the common ®nal step. To date, the most widely used approach has been to separate protein samples by 2D gel electrophoresis. However, this method does not allow the preservation of proteins as complexes. Alternative techniques have been developed to avoid this problem, allowing researchers to discover which proteins exist as complexes with other proteins; the advantages of these approaches as compared with 2D gel electrophoresis are discussed in the review. Beyond the quanti®cation of each protein present in the cell lies the determination of the interactions between proteins and their state of post-trans-lational modi®cation. In this issue, we highlight the development of the two-hybrid protein interaction screen into a genome-wide technique. The role of global two-hybrid interaction screens in proteomics is discussed by Albertha Walhout, Simon Boulton and Marc Vidal. Two-hybrid screening is an established approach for identifying protein interaction partners. In this review, the advance into whole-genomèglobal' two-hybrid screens is charted by discussion of the recent large-scale studies in Saccharomyces cerevisiae and Caenorhabditis elegans. The various approaches that have been used so far are compared with each other and the advantages and drawbacks of each are considered. Several other issues, such as how the data being generated by these approaches should be assessed and the average number ofìnteraction sequence tags' per protein to be expected from such a screen, are also discussed. Global two-hybrid screen in yeast yields new splicing factors Fromont-Racine et al. have used a genome wide two-hybrid screen with yeast genes encoding proteins thought to be involved in splicing as baits. The Sm proteins form the core of the snRNP particles that contain the spliceosomal snRNAs. A family of Sm-like (Lsm) proteins were previously detected by the authors in sequence homology searches of the yeast genome. These were used as baits in exhaustive two-hybrid screens with a mixture of whole and domain-encoding regions of all yeast genes as prey. …