Advertisement

Mining from Complete Proteomes to Identify Adhesins and Adhesin-Like Proteins: A Rapid Aid to Experimental Researchers

  • S. RamachandranEmail author
  • P. Jain
  • K. Kumar
  • G. Sachdeva
Chapter
  • 744 Downloads

Abstract

Adhesins are microbial surface proteins that mediate the adherence of microbial pathogens to host cell surfaces. This interaction is often the first step in the establishment of a disease. Identification of novel adhesins and their characterization are important for studying host-pathogen interactions and for testing new vaccine formulations prepared from adhesins. Currently, experimental methods are used for detecting and characterizing adhesins, which is a time-consuming task and demands large resources. The availability of a software program specifically focused to identifying adhesins from the predicted proteomes of microbial pathogens can aid experimenters in simplifying the complexity of this problem. We have employed artificial neural networks to develop an algorithm SPAAN, which predicts the probability of a protein being an adhesin Pad based on 105 compositional properties computed from its sequence. SPAAN had optimal sensitivity of 89 % and specificity of 100 % on a defined test data set and could identify 97.4 % of the known adhesins at a high Pad value from a wide range of bacteria. Data mining using SPAAN not only identified the known adhesins, but also guided in improvement of annotation of several proteins as adhesins. Several novel adhesins were identified in many pathogenic organisms causing diseases in humans and plants. These results offer new leads for rapid experimental testing.

Key words

virulence factors adhesins vaccine neural networks 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media, Inc. 2006

Authors and Affiliations

  • S. Ramachandran
    • 1
    Email author
  • P. Jain
    • 1
  • K. Kumar
    • 1
  • G. Sachdeva
    • 1
  1. 1.G.N. Ramachandran Knowledge Center for Genome InformaticsInstitute of Genomics and Integrative BiologyDelhiIndia

Personalised recommendations