Optimization-based inference for temporally evolving networks with applications in biology

J Comput Biol. 2012 Dec;19(12):1307-23. doi: 10.1089/cmb.2012.0190.

Abstract

The problem of identifying dynamics of biological networks is of critical importance in order to understand biological systems. In this article, we propose a data-driven inference scheme to identify temporally evolving network representations of genetic networks. In the formulation of the optimization problem, we use an adjacency map as a priori information and define a cost function that both drives the connectivity of the graph to match biological data as well as generates a sparse and robust network at corresponding time intervals. Through simulation studies of simple examples, it is shown that this optimization scheme can help capture the topological change of a biological signaling pathway, and furthermore, might help to understand the structure and dynamics of biological genetic networks.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Algorithms*
  • Breast Neoplasms
  • Computational Biology / methods*
  • Computer Simulation
  • Female
  • Gene Regulatory Networks*
  • Humans
  • Models, Biological*
  • Nonlinear Dynamics
  • Receptor, ErbB-2 / metabolism
  • Signal Transduction*

Substances

  • Receptor, ErbB-2