Regensburg 2013 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 32: Statistical Physics in Biological Systems III (joint with BP)

DY 32.7: Talk

Thursday, March 14, 2013, 16:30–16:45, H46

Score statistics of multiple sequence alignments — •Pascal Fieth and Alexander K. Hartmann — Institute of Physics, University of Oldenburg

Optimally aligned sequences of amino acids [1] can be studied numerically [2] in the biologically relevant high scoring region by means of parallel tempering simulations [3]. Preceding studies have shown that the scores of gapped pairwise sequence alignments of finite-length sequences follow a Gumbel extreme value distribution, modified by a Gaussian correction [4] rather than a simple Gumbel extreme value distribution as previously predicted for ungapped pairwise alignments. In this study these methods are applied to the case of multiple sequence alignment (MSA). Here the distributions of the sum-of-pair scores of the MSA of more than two sequences are studied. In particular the distribution of protein MSA-scores using different common substitution matrices (BLOSUM and PAM) are analysed for protein background frequencies of real sequences.

[1] R. Durbin et al., Biological sequence analysis, (Cambridge University Press, 1998)

[2] A.K. Hartmann, Practical Guide to Computer Simulations, (World Scientific, 2009)

[3] A.K. Hartmann and Heiko Rieger, Optimization Algorithms in Physics, (Wiley-VCH, 2001)

[4] S. Wolfsheimer et al., Local Sequence Alignments Statistics: Deviations from the Gumbel Statistics in the Rare-Event Tail, (Algorithms for Molecular Biology, 2007)