A two-sample Bayesian t-test for microarray data.

BACKGROUND: Determining whether a gene is differentially expressed in two different samples remains an important statistical problem. Prior work in this area has featured the use of t-tests with pooled estimates of the sample variance based on similarly expressed genes. These methods do not display consistent behavior across the entire range of pooling and can be biased when the prior hyperparameters are specified heuristically. RESULTS: A two-sample Bayesian t-test is proposed for use in determining whether a gene is differentially expressed in two different samples. The test method is an extension of earlier work that made use of point estimates for the variance. The method proposed here explicitly calculates in analytic form the marginal distribution for the difference in the mean expression of two samples, obviating the need for point estimates of the variance without recourse to posterior simulation. The prior distribution involves a single hyperparameter that can be calculated in a statistically rigorous manner, making clear the connection between the prior degrees of freedom and prior variance. CONCLUSION: The test is easy to understand and implement and application to both real and simulated data shows that the method has equal or greater power compared to the previous method and demonstrates consistent Type I error rates. The test is generally applicable outside the microarray field to any situation where prior information about the variance is available and is not limited to cases where estimates of the variance are based on many similar observations.

Draft genome of the filarial nematode parasite Brugia malayi.

Parasitic nematodes that cause elephantiasis and river blindness threaten hundreds of millions of people in the developing world. We have sequenced the approximately 90 megabase (Mb) genome of the human filarial parasite Brugia malayi and predict approximately 11,500 protein coding genes in 71 Mb of robustly assembled sequence. Comparative analysis with the free-living, model nematode Caenorhabditis elegans revealed that, despite these genes having maintained little conservation of local synteny during approximately 350 million years of evolution, they largely remain in linkage on chromosomal units. More than 100 conserved operons were identified. Analysis of the predicted proteome provides evidence for adaptations of B. malayi to niches in its human and vector hosts and insights into the molecular basis of a mutualistic relationship with its Wolbachia endosymbiont. These findings offer a foundation for rational drug design.

Detecting coevolving amino acid sites using Bayesian mutational mapping.

MOTIVATION: The evolution of protein sequences is constrained by complex interactions between amino acid residues. Because harmful substitutions may be compensated for by other substitutions at neighboring sites, residues can coevolve. We describe a Bayesian phylogenetic approach to the detection of coevolving residues in protein families. This method, Bayesian mutational mapping (BMM), assigns mutations to the branches of the evolutionary tree stochastically, and then test statistics are calculated to determine whether a coevolutionary signal exists in the mapping. Posterior predictive P-values provide an estimate of significance, and specificity is maintained by integrating over uncertainty in the estimation of the tree topology, branch lengths and substitution rates. A coevolutionary Markov model for codon substitution is also described, and this model is used as the basis of several test statistics. RESULTS: Results on simulated coevolutionary data indicate that the BMM method can successfully detect nearly all coevolving sites when the model has been correctly specified, and that non-parametric statistics such as mutual information are generally less powerful than parametric statistics. On a dataset of eukaryotic proteins from the phosphoglycerate kinase (PGK) family, interdomain site contacts yield a significantly greater coevolutionary signal than interdomain non-contacts, an indication that the method provides information about interacting sites. Failure to account for the heterogeneity in rates across sites in PGK resulted in a less discriminating test, yielding a marked increase in the number of reported positives at both contact and non-contact sites. SUPPLEMENTARY INFORMATION: http://www.dimmic.net/supplement/

Dimerization in aminergic G-protein-coupled receptors: application of a hidden-site class model of evolution.

G-Protein-coupled receptors (GPCRs) are an important superfamily of transmembrane proteins involved in cellular communication. Recently, it has been shown that dimerization is a widely occurring phenomenon in the GPCR superfamily, with likely important physiological roles. Here we use a novel hidden-site class model of evolution as a sequence analysis tool to predict possible dimerization interfaces in GPCRs. This model aims to simulate the evolution of proteins at the amino acid level, allowing the analysis of their sequences in an explicitly evolutionary context. Applying this model to aminergic GPCR sequences, we first validate the general reasoning behind the model. We then use the model to perform a family specific analysis of GPCRs. Accounting for the family structure of these proteins, this approach detects different evolutionarily conserved and accessible patches on transmembrane (TM) helices 4-6 in different families. On the basis of these findings, we propose an experimentally testable dimerization mechanism, involving interactions among different combinations of these helices in different families of aminergic GPCRs.

rtREV: an amino acid substitution matrix for inference of retrovirus and reverse transcriptase phylogeny.

Retroviral and other reverse transcriptase (RT)-containing sequences may be subject to unique evolutionary pressures, and models of molecular sequence evolution developed using other kinds of sequences may not be optimal. Here we develop and present a new substitution matrix for maximum likelihood (ML) phylogenetic analysis which has been optimized on a dataset of 33 amino acid sequences from the retroviral Pol proteins. When compared to other matrices, this model (rtREV) yields higher log-likelihood values on a range of datasets including lentiviruses, spumaviruses, betaretroviruses, gammaretroviruses, and other elements containing reverse transcriptase. We provide evidence that rtREV is a more realistic evolutionary model for analyses of the pol gene, although it is inapplicable to analyses involving the gag gene.

Using evolutionary methods to study G-protein coupled receptors.

A novel method to analyze evolutionary change is presented and its application to the analysis of sequence data is discussed. The investigated method uses phylogenetic trees of related proteins with an evolutionary model in order to gain insight about protein structure and function. The evolutionary model, based on amino acid substitutions, contains adjustable parameters related to amino acid and sequence properties. A maximum likelihood approach is used with a phylogenetic tree to optimize these parameters. The model is applied to a set of Muscarinic receptors, members of the G-protein coupled receptor family. Here we show that the optimized parameters of the model are able to highlight the general structural features of these receptors.