Graph Kernels for Disease Outcome Prediction from Protein-Protein Interaction Networks

K. M. Borgwardt, H. Kriegel, S. Vishwanathan, and N. N. Schraudolph. Graph Kernels for Disease Outcome Prediction from Protein-Protein Interaction Networks. In Proc. Pacific Symposium on Biocomputing (PSB), pp. 4–15, World Scientific, Maui, Hawaii, 2007.

Download

179.9kB	81.4kB	1.2MB

Abstract

It is widely believed that comparing discrepancies in the protein-protein interaction (PPI) networks of individuals will become an important tool in understanding and preventing diseases. Currently PPI networks for individuals are not available, but gene expression data is becoming easier to obtain and allows us to represent individuals by a co-integrated gene expression/protein interaction network. Two major problems hamper the application of graph kernels---state-of-the-art methods for whole-graph comparison---to compare PPI networks. First, these methods do not scale to graphs of the size of a PPI network. Second, missing edges in these interaction networks are biologically relevant for detecting discrepancies, yet, these methods do not take this into account. In this article we present graph kernels for biological network comparison that are fast to compute and take into account missing interactions. We evaluate their practical performance on two datasets of co-integrated gene expression/PPI networks.

BibTeX Entry

@inproceedings{BorKriVisSch07,
     author = {Karsten M. Borgwardt and Hans-Peter Kriegel and
               S.~V.~N. Vishwanathan and Nicol N. Schraudolph},
      title = {\href{http://nic.schraudolph.org/pubs/BorKriVisSch07.pdf}{
               Graph Kernels for Disease Outcome Prediction
               from Protein-Protein Interaction Networks}},
      pages = {4--15},
  booktitle = {Proc.\ Pacific Symposium on Biocomputing (PSB)},
     volume =  12,
  publisher = {World Scientific},
       isbn = {978-981-270-417-7},
    address = {Maui, Hawaii},
       year =  2007,
   b2h_type = {Top Conferences},
  b2h_topic = {Kernel Methods, Bioinformatics},
   abstract = {
    It is widely believed that comparing discrepancies in the
    protein-protein interaction (PPI) networks of individuals will
    become an important tool in understanding and preventing diseases.
    Currently PPI networks for individuals are not available, but
    gene expression data is becoming easier to obtain and allows
    us to represent individuals by a co-integrated gene expression/protein
    interaction network. Two major problems hamper the application
    of graph kernels\,---\,state-of-the-art methods for whole-graph
    comparison\,---\,to compare PPI networks. First, these methods
    do not scale to graphs of the size of a PPI network. Second,
    missing edges in these interaction networks are biologically
    relevant for detecting discrepancies, yet, these methods do not
    take this into account. In this article we present graph kernels
    for biological network comparison that are fast to compute and
    take into account missing interactions. We evaluate their
    practical performance on two datasets of co-integrated gene
    expression/PPI networks.
}}