Conference Paper

Protein interaction prediction for mouse pdz domains using dipeptide composition features

Key Lab. of Syst. Biol., Chinese Acad. of Sci., Shanghai, China
DOI: 10.1109/ISB.2011.6033143 Conference: Systems Biology (ISB), 2011 IEEE International Conference on
Source: IEEE Xplore

ABSTRACT The PDZ domain is one of the largest families of protein domains that are involved in targeting and routing specific proteins in signaling pathways. PDZ domains mediate protein-protein interactions by binding the C-terminal peptides of their target proteins. Using the dipeptide feature encoding, we develop a PDZ domain interaction predictor using a support vector machine that achieves a high accuracy rate of 82.49%. Since most of the dipeptide compositions are redundant and irrelevant, we propose a new hybrid feature selection technique to select only a subset of these compositions that are useful for interaction prediction. Our experimental results show that only approximately 25% of dipeptide features are needed and that our method increases the accuracy by 3%. The selected dipeptide features are analyzed and shown to have important roles on specificity pattern of PDZ domains.

0 Bookmarks
 · 
95 Views
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Modulating protein-protein interactions involved in disease pathways is an attractive strategy for developing drugs, but remains a challenge to achieve. One approach is to target certain domains within proteins that mediate these interactions. One example of such a domain is the PDZ domain, which is involved in interactions between many different proteins in a variety of cellular contexts. Because PDZ domains have well-defined binding sites, they are promising targets for drug discovery. However, there is still much to learn about the function of these domains before drugs targeting PDZ interactions can become a reality.
    dressNature Reviews Drug Discovery 01/2005; 3(12):1047-56. · 33.08 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The N-methyl-D-aspartate (NMDA) receptor subserves synaptic glutamate-induced transmission and plasticity in central neurons. The yeast two-hybrid system was used to show that the cytoplasmic tails of NMDA receptor subunits interact with a prominent postsynaptic density protein PSD-95. The second PDZ domain in PSD-95 binds to the seven-amino acid, COOH-terminal domain containing the terminal tSXV motif (where S is serine, X is any amino acid, and V is valine) common to NR2 subunits and certain NR1 splice forms. Transcripts encoding PSD-95 are expressed in a pattern similar to that of NMDA receptors, and the NR2B subunit co-localizes with PSD-95 in cultured rat hippocampal neurons. The interaction of these proteins may affect the plasticity of excitatory synapses.
    Science 10/1995; 269(5231):1737-40. · 31.03 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Syntrophins are modular proteins belonging to the dystrophin associated glycoprotein complex and are thought to be involved in the regulation of the muscular system. Screening of peptide libraries revealed selectivity of the synotrophin PDZ domain toward the motif R/K/Q-E-S/T-X-V-COO- found to be highly conserved in the alpha-subunit C-terminus of vertebrate voltage gated sodium channels (VGSCs). The solution structure of the domain in complex with the peptide G-V-K-E-S-L-V shows specific interactions between the conserved residues in the peptide and syntrophin-characteristic residues in the domain. We propose that syntrophins localize VGSCs to the dystrophin network through its PDZ domain.
    Nature Structural Biology 02/1998; 5(1):19-24.