IMA Journal of Numerical Analysis Advance Access published online on May 2, 2008
IMA Journal of Numerical Analysis, doi:10.1093/imanum/drn012
Trust-region superposition methods for protein alignment


Department of Applied Mathematics, IMECC-UNICAMP, State University of Campinas, CP 6065, 13081-970 Campinas SP, Brazil

Institute of Chemistry, State University of Campinas, SP, Brazil and Institut Pasteur, Paris, France
Email: andreani{at}ime.unicamp.br
Corresponding author. Email: martinez{at}ime.unicamp.br
Email: lmartinez{at}iqm.unicamp.br
Received on 31 January 2007. Accepted for publication 1 February 2008.
| Abstract |
|---|
Protein alignment is a challenging applied optimization problem. Superposition methods are based on the maximization of a score function with respect to rigid-body modifications of relative positions. The problem of score maximization can be modelled as a continuous nonsmooth optimization problem (low order-value optimization (LOVO)). This allows one to define practical and convergent methods that produce monotone increases of the score. In this paper, trust-region methods are introduced for solving the problem. Numerical results are presented. Computer software related to the LOVO approach for protein alignment is available at www.ime.unicamp.br/
martinez/lovoalign.
Key Words: protein alignment; trust-region methods; low order-value optimization
This work is dedicated to the 70th birthday of Prof. M. J. D. Powell in September 2006.