Rosetta Stone proteins chance and necessity 英文参考文献.docVIP

/2002/3/2/interactions/1001.1 Correspondence Rosetta Stone proteins: “chance and necessity”? Reiner A Veitia A response to Functional associations of proteins in entire genomes by means of exhaustive detection of gene fusions by AJ Enright, CA Ouzounis. Genome Biology 2001, 2:research0034.1-0034.7 Address: UFR de Biologie et Sciences de la Nature, Université Denis Diderot/Paris VII, Immunogénétique Humaine, Institut Pasteur, 75724 Paris, France. E-mail: rveitia@pasteur.fr Published: 8 January 2002 GenomeBiology 2002, 3(2):interactions1001.1–1001.3 The electronic version of this article is the complete one and can be found online at /2002/3/2/interactions/1001 ? BioMed Central Ltd (Print ISSN 1465-6906; Online ISSN 1465-6914) The field of predicting protein-protein interactions has been active for over two decades, but in silico methods have become prominent in the last three years with the expansion of analyses at a genomic scale [1,2]. The rationale of one of the computational methods is as simple as it is elegant. Two polypep- tides A and B in one organism are likely to interact if their homologs are expressed as a single polypeptide AB in another. The latter polypeptide (AB) is called a Rosetta Stone protein, as it contains information about both A and B. Marcotte et al. [1] have proposed that fusion to form a single polypep- tide reduces the entropy of dissocia- tion of A and B. The result is a huge increase in the local concentration of A with respect to B. A recent paper in Genome Biology describes the effort of Enright and Ouzounis [3], who carried out a vast analysis of genes of the Rosetta Stone type over 24 genomes, including eukaryotic genomes. They uncovered many new ‘composite’ or ‘Rosetta’ proteins, many of them con- tributed by eukaryotes. Here, I provide related genes are clustered. Dandekar et al. [4] compared nine