Context-dependent optimal substitution matrices

Abstract

Substitution matrices are a key tool in important applications such as identifying sequence homologies, creating sequence alignments and more recently using evolutionary patterns for the prediction of protein structure. We have derived a novel approach to the derivation of these matrices that utilizes not only multiple sequence alignments, but also the associated evolutionary trees. The key to our method is the use of a Bayesian formalism to calculate the probability that a given substitution matrix fits the tree structures and multiple sequence alignment data. Using this procedure, we can determine optimal substitution matrices for various local environments, depending on parameters such as secondary structure and surface accessibility.

Site interdependence attributed to tertiary structure in amino acid sequence evolution

“…Models with sitespecific matrices have also been studied (Bruno, 1996;Halpern and Bruno, 1998), as well as various types of mixture models (e.g. Koshi and Goldstein, 1995;Thorne et al, 1996;Koshi and Goldstein, 1997;Goldman et al, 1998;Lartillot and Philippe, 2004;Pagel and Meade, 2004).…”

Real value prediction of protein folding rate change upon point mutation

“…First, we trained our model with all the 621 mutants using 173 amino acid properties obtained from AAindex database. These properties have been examined using genetic algorithm (details are given in supplementary material) and extracted a set of 7 properties, which are given below: amino acid properties (1) wild type pHi, (isoelectric point), context-dependent optimal substitution matrices [67] and (5) structure-conservation scoring tables [68]; the difference of statistical contact potentials in contact with the first neighboring residue along the (6) N-terminus [69] and (7) C-terminus [70]. The 10-fold and 20-fold cross-validation accuracies for C559 and C559 ?…”

Molecular Structure and Evolution of Genomes

“…The principal difficulty in modeling protein evolution is that it is highly context dependent, meaning that the probability of amino acid residue replacement during evolution is expected to vary across positions and over time (Andreeva and Murzin, 2006;Buchner, 1999;Koshi and Goldstein, 1995;Kozak, 1999;Midelfort and Wittrup, 2006;Pollock and Goldstein, 2002;Templeton et al, 2004;Xu et al, 2005). Particular positions in a protein will have different structural and functional environments, and thus different evolutionary constraints, and will therefore have distinctive patterns of substitution at the corresponding codons.…”