The Bayesian mixture model CAT [ 64] implemented in PhyloBayes 3.3f [ 36, 37] was used on the concatenated 172 amino acid alignment to partition sites into different rate categories using non-parametric modeling of site specific effects.
As described above, in the Codon Sorting section, we have used structural information to partition sites into classes, which are expected to have different selective pressures and thus different ω ratios.
Therefore, as a first step, we have explored the influence of the partitioning site of an aromatic salt counterion near the gemini micelle and its possible impact on overall micellar structural changes.
Traditional data partitioning schemes try to account for variation in evolutionary rates by partitioning nucleotide sites based on their rate of evolution, most commonly, by gene or codon position.
Partitioning the sites into groups according to ancestral amino acid identity explains some of the variation, with mutations at certain amino acids being unconditionally deleterious, others unconditionally nearly neutral, and others with substantially variable effects.
A second proposed advantage of BI over ML is that Bayesian MCMC provides a more reliable method for analyzing data using complex models that incorporate evolutionary heterogeneity among sites, such as those that use mixture models or partition sites into independent classes [3], [10], [12], [14].
Histone deacetylation sites may serve as possible partition sites for chromatin domains and affect gene expression.
