Discovering transcriptional modules by Bayesian data integration.

MOTIVATION: We present a method for directly inferring transcriptional modules (TMs) by integrating gene expression and transcription factor binding (ChIP-chip) data. Our model extends a hierarchical Dirichlet process mixture model to allow data fusion on a gene-by-gene basis. This encodes the intuition that co-expression and co-regulation are not necessarily equivalent and hence we do not expect all genes to group similarly in both datasets. In particular, it allows us to identify the subset of genes that share the same structure of transcriptional modules in both datasets. RESULTS: We find that by working on a gene-by-gene basis, our model is able to extract clusters with greater functional coherence than existing methods. By combining gene expression and transcription factor binding (ChIP-chip) data in this way, we are better able to determine the groups of genes that are most likely to represent underlying TMs. AVAILABILITY: If interested in the code for the work presented in this article, please contact the authors. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Mxr1p, a key regulator of the methanol utilization pathway and peroxisomal genes in Pichia pastoris.

Growth of the yeast Pichia pastoris on methanol induces the expression of genes whose products are required for its metabolism. Three of the methanol pathway enzymes are located in an organelle called the peroxisome. As a result, both methanol pathway enzymes and proteins involved in peroxisome biogenesis (PEX proteins) are induced in response to this substrate. The most highly regulated of these genes is AOX1, which encodes alcohol oxidase, the first enzyme of the methanol pathway, and a peroxisomal enzyme. To elucidate the molecular mechanisms responsible for methanol regulation, we identify genes required for the expression of AOX1. Mutations in one gene, named MXR1 (methanol expression regulator 1), result in strains that are unable to (i) grow on the peroxisomal substrates methanol and oleic acid, (ii) induce the transcription of AOX1 and other methanol pathway and PEX genes, and (iii) form normal-appearing peroxisomes in response to methanol. MXR1 encodes a large protein with a zinc finger DNA-binding domain near its N terminus that has similarity to Saccharomyces cerevisiae Adr1p. In addition, Mxr1p is localized to the nucleus in cells grown on methanol or other gluconeogenic substrates. Finally, Mxr1p specifically binds to sequences upstream of AOX1. We conclude that Mxr1p is a transcription factor that is necessary for the activation of many genes in response to methanol. We propose that MXR1 is the P. pastoris homologue of S. cerevisiae ADR1 but that it has gained new functions and lost others through evolution as a result of changes in the spectrum of genes that it controls.

Lipid composition of peroxisomes from the yeast Pichia pastoris grown on different carbon sources.

Highly purified peroxisomes from the yeast Pichia pastoris grown on methanol or oleic acid, respectively, were used to characterize the lipid composition of this organelle. For this purpose, an isolation procedure had to be adapted which yielded highly purified P. pastoris peroxisomes. When peroxisome proliferation was induced by growth on methanol, alcohol oxidase was the predominant peroxisomal protein. Cultivation of P. pastoris on oleic acid led to induction of a family of peroxisomal enzymes catalyzing fatty acid beta-oxidation, whose most prominent members were identified by mass spectrometry. On either carbon source, phosphatidylcholine and phosphatidylethanolamine were the major peroxisomal phospholipids, and cardiolipin was present in peroxisomal membranes at a substantial amount, indicating that this phospholipid is a true peroxisomal component. Ergosterol was the most abundant sterol of P. pastoris peroxisomal membranes irrespective of the culture conditions. The fatty acid composition of whole cells and peroxisomes was highly affected by cultivation of P. pastoris on oleic acid. Under these conditions, oleic acid became the predominant fatty acid in phospholipids from total cell and peroxisomal extracts. Thus, oleic acid was not only utilized as an appropriate carbon source but also as a building block for complex membrane lipids. In summary, our data provide first insight into biochemical properties of P. pastoris peroxisomal membranes, which may become important for the biotechnological use of this yeast.

Modeling and visualizing uncertainty in gene expression clusters using dirichlet process mixtures.

Although the use of clustering methods has rapidly become one of the standard computational approaches in the literature of microarray gene expression data, little attention has been paid to uncertainty in the results obtained. Dirichlet process mixture (DPM) models provide a nonparametric Bayesian alternative to the bootstrap approach to modeling uncertainty in gene expression clustering. Most previously published applications of Bayesian model-based clustering methods have been to short time series data. In this paper, we present a case study of the application of nonparametric Bayesian clustering methods to the clustering of high-dimensional nontime series gene expression data using full Gaussian covariances. We use the probability that two genes belong to the same cluster in a DPM model as a measure of the similarity of these gene expression profiles. Conversely, this probability can be used to define a dissimilarity measure, which, for the purposes of visualization, can be input to one of the standard linkage algorithms used for hierarchical clustering. Biologically plausible results are obtained from the Rosetta compendium of expression profiles which extend previously published cluster analyses of this data.

The role of the 5' untranslated region (UTR) in glucose-dependent mRNA decay.

When S. cerevisiae are grown with glucose, SDH2 mRNA encoding the iron protein of the succinate dehydrogenase complex is unstable and present at low level. In yeast grown without glucose, SDH2 mRNA is stable and its level rises. Addition of glucose to a glucose-limited culture causes the SDH2 mRNA level to fall rapidly with a half-life of approximately 5-7 min. Previously the 5'UTR of the mRNA of SDH2 was shown to be necessary and sufficient to destabilize it in glucose (Lombardo et al., 1992). We now show that the SDH1 and SUC2 5'UTRs are capable of conferring glucose-sensitive mRNA instability. We also examine how changes in the SDH2 5'UTR affect glucose-triggered degradation. Finally, we show that changes in mRNA stability are correlated with changes in translational efficiency for these transcripts.