Improved statistical inference from DNA microarray data using analysis of variance and a Bayesian statistical framework. Analysis of global gene expression in Escherichia coli K12

We describe statistical methods based on the t test that can be conveniently used on high density array data to test for statistically significant differences between treatments. These t tests employ either the observed variance among replicates within treatments or a Bayesian estimate of the variance among replicates within treatments based on a prior estimate obtained from a local estimate of the standard deviation. The Bayesian prior allows statistical inference to be made from microarray data even when experiments are only replicated at nominal levels. We apply these new statistical tests to a data set that examined differential gene expression patterns in IHF(+) and IHF(-) Escherichia coli cells (Arfin, S. M., Long, A. D., Ito, E. T., Tolleri, L., Riehle, M. M., Paegle, E. S., and Hatfield, G. W. (2000) J. Biol. Chem. 275, 29672-29684). These analyses identify a more biologically reasonable set of candidate genes than those identified using statistical tests not incorporating a Bayesian prior. We also show that statistical tests based on analysis of variance and a Bayesian prior identify genes that are up- or down-regulated following an experimental manipulation more reliably than approaches based only on a t test or fold change. All the described tests are implemented in a simple-to-use web interface called Cyber-T that is located on the University of California at Irvine genomics web site.     

The role of endocytosis in regulating L1-mediated adhesion

L1 is a neural cell adhesion molecule critical for neural development. Full-length L1 (L1(FL)) contains an alternatively spliced cytoplasmic sequence, RSLE, which is absent in L1 expressed in nonneuronal cells. The RSLE sequence follows a tyrosine, creating an endocytic motif that allows rapid internalization via clathrin-mediated endocytosis. We hypothesized that L1(FL) would internalize more rapidly than L1 lacking the RSLE sequence (L1(Delta)(RSLE)) and that internalization might regulate L1-mediated adhesion. L1 internalization was measured by immunofluorescence microscopy and by uptake of (125)I-anti-rat-L1 antibody, demonstrating that L1(FL) is internalized 2-3 times faster than L1(Delta)(RSLE). Inhibition of clathrin-mediated endocytosis slowed internalization of L1(FL) but did not affect initial uptake of L1(Delta)(RSLE). To test whether L1 endocytosis regulates L1 adhesion, cell aggregation rates were tested. L1(Delta)(RSLE) cells aggregated two times faster than L1(FL) cells. Inhibition of clathrin-mediated endocytosis increases the aggregation rate of the L1(FL) cells to that of L1(Delta)(RSLE) cells. Our results demonstrate that rapid internalization of L1 dramatically affects L1 adhesion.     

A Bayesian framework for the analysis of microarray expression data: regularized t -test and statistical inferences of gene changes

MOTIVATION: DNA microarrays are now capable of providing genome-wide patterns of gene expression across many different conditions. The first level of analysis of these patterns requires determining whether observed differences in expression are significant or not. Current methods are unsatisfactory due to the lack of a systematic framework that can accommodate noise, variability, and low replication often typical of microarray data. RESULTS: We develop a Bayesian probabilistic framework for microarray data analysis. At the simplest level, we model log-expression values by independent normal distributions, parameterized by corresponding means and variances with hierarchical prior distributions. We derive point estimates for both parameters and hyperparameters, and regularized expressions for the variance of each gene by combining the empirical variance with a local background variance associated with neighboring genes. An additional hyperparameter, inversely related to the number of empirical observations, determines the strength of the background variance. Simulations show that these point estimates, combined with a t -test, provide a systematic inference approach that compares favorably with simple t -test or fold methods, and partly compensate for the lack of replication.     

Provenance Of Atlantic Lingulid Brachiopods

Living lingulid brachiopods are ubiquitous in low-latitude, marine infaunas. Lingula occurs throughout the Pacific and Indian oceans with the only Atlantic species, L. parva, confined to West Africa. Glottidia is restricted to offshore America from Virginia to California and Peru, and is assumed to have descended from a Pacific Lingula during the early Tertiary. Lingulid organophosphatic shells differ structurally. That of Glottidia is characterizedby trellised rods (baculate); that of Indo-Pacific species of Lingula by spheroidal and rod-like microstructures (virgose); and that of L. parva by apatitic rods arranged as spherulites. A spherulitic fabric is unknown in fossil lingulids, but the distinction between GlottidiaLingula can be traced back to the Carboniferous, which accords with the deep molecular divergence between the two genera. The common occurrence of lingulids with baculate shells in European post-Palaeozoic sediments suggests that ancestral Glottidia entered the Atlantic by the Tethyan Current during the Late Cretaceous/early Cenozoic, and migrated into the Pacific before the formation of the Panama Isthmus. Penecontemporaneously, antecedents of L. parva possibly migrated from east Tethys along the trans-Saharan seaway.     

ExInt: an Exon/Intron database

The Exon/Intron (ExInt) database incorporates information on the exon/intron structure of eukaryotic genes. Features in the database include: intron nucleotide sequence, amino acid sequence of the corresponding protein, position of the introns at the amino acid level and intron phase. From ExInt, we have also generated four additional databases each with ExInt entries containing predicted introns, introns experimentally defined, organelle introns or nuclear introns. ExInt is accessible through a retrieval system with pointers to GenBank. The database can be searched by keywords, locus name, NID, accession number or length of the protein. ExInt is freely accessible at http://intron.bic.nus.edu.sg/exint/exint.html     

Cerebral artery sarcoplasmic reticulum Ca(2+) stores and contractility: changes with development

To test the hypothesis that sarcoplasmic reticulum (SR) Ca(2+) stores play a key role in norepinephrine (NE)-induced contraction of fetal and adult cerebral arteries and that Ca(2+) stores change with development, we performed the following study. In main branch middle cerebral arteries (MCA) from near-term fetal ( approximately 140 days) and nonpregnant adult sheep, we measured NE-induced contraction and intracellular Ca(2+) concentration ([Ca(2+)](i)) in the absence and presence of different blockers. In adult MCA, after thapsigargin (10(-6) M), the NE-induced responses of tension and [Ca(2+)](i) were 37 +/- 5 and 47 +/- 7%, respectively, of control values (P < 0.01 for each). In the fetal artery, in contrast, this treatment resulted in no significant changes from control. When this was repeated in the absence of extracellular Ca(2+), adult MCA increases in tension and [Ca(2+)](i) were 32 +/- 5 and 13 +/- 3%, respectively, of control. Fetal cerebral arteries, however, showed essentially no response. Ryanodine (RYN, 3 x 10(-6) to 10(-5) M) resulted in increases in tension and [Ca(2+)](i) in both fetal and adult MCA similar to that seen with NE. For both adult and fetal MCA, the increased tension and [Ca(2+)](i) responses to RYN were essentially eliminated in the presence of zero extracellular Ca(2+). These findings provide evidence that in fetal MCA, in contrast to those in the adult, SR Ca(2+) stores are of less importance in NE-induced contraction, with such contraction being almost wholly dependent on Ca(2+) flux via plasma membrane L-type Ca(2+) channels. In addition, they suggest that in both adult and fetal MCA, the RYN receptor is coupled to the plasma membrane Ca(2+)-activated K(+) channel and/or L-type Ca(2+) channel.     

PHTS, a novel putative tumor suppressor, is involved in the transformation reversion of HeLaHF cells independently of the p53 pathway

HeLaHF is a non-transformed revertant of HeLa cells, likely resulting from the activation of a putative tumor suppressor(s). p53 protein was stabilized in this revertant and reactivated for certain transactivation functions. Although p53 stabilization has not conclusively been linked to the reversion, it is clear that the genes in p53 pathway are involved. The present study confirms the direct role of p53 in HeLaHF reversion by demonstrating that RNAi-mediated p53 silencing partially restores anchorage-independent growth potential of the revertant through the suppression of anoikis. In addition, we identified a novel gene, named PHTS, with putative tumor suppressor properties, and showed that this gene is also involved in HeLaHF reversion independently of the p53 pathway. Expression profiling revealed that PHTS is one of the genes that is up-regulated in HeLaHF but not in HeLa. It encodes a putative protein with CD59-like domains. RNAi-mediated PHTS silencing resulted in the partial restoration of transformation (anchorage-independent growth) in HeLaHF cells, similar to that of p53 gene silencing, implying its tumor suppressor effect. However, the observed increased transformation potential by PHTS silencing appears to be due to an increased anchorage-independent proliferation rate rather than suppression of anoikis, unlike the effect of p53 silencing. p53 silencing did not affect PHTS gene expression, and vice versa, suggesting PHTS may function in a new and p53-independent tumor suppressor pathway. Furthermore, over-expression of PHTS in different cancer cell lines, in addition to HeLa, reduces cell growth likely via induced apoptosis, confirming the broad PHTS tumor suppressor properties.