A new, structurally nonredundant, diverse data set of protein–protein interfaces and its implications

• DOI: 10.1110/ps.03484604
• Publication Date: 2004-03-25T18:44:54Z
• Authors: Ozlem Keskin; Chung‐Jung Tsai; Haim Wolfson; Ruth Nussinov
• Keywords: Interface (matter); Hierarchical clustering; Sequence (biology)

Here, we present a diverse, structurally nonredundant data set of two-chain protein-protein interfaces derived from the PDB. Using sequence order-independent structural comparison algorithm and hierarchical clustering, 3799 interface clusters are obtained. These yield 103 with at least five nonhomologous members. We divide into three types. In Type I clusters, global structures chains which also similar. This cluster type is expected because, in general, related proteins associate similar ways. II, similar; however, remarkably, overall functions different. The functional spectrum broad, enzymes/inhibitors to immunoglobulins toxins. fact that different monomers ways, suggests "good" binding architectures. observation extends paradigm protein science: It has been well known may have functions. show it interfaces. III only one side across cluster. provides rich for studies interactions recognition, cellular networks drug design. particular, be useful addressing difficult question what favorable ways interact. (The available http://protein3d.ncifcrf.gov/~keskino/ http://home.ku.edu.tr/~okeskin/INTERFACE/INTERFACES.html.)