Functional RNA sequences typically have structural elements that are highly conserved during evolution. Here we present an algorithmic method for multiple alignment of RNAs, taking into consideration both structural similarity and sequence identity. Furthermore, we performed a comparative analysis on pairing probability matrices of a set of aligned orthologous sequences and predicted the conserved secondary structure. Our alignment method outperforms the most widely used multiple alignment tool - Clustal W, and the structure prediction approach we proposed can generate a more accurate secondary structure for 5S rRNA compared to the existing approaches such as Alifold. In addition, our algorithms are efficient in terms of CPU time and memory usage compared to most existing methods for secondary structure prediction.

Regulators of G-protein coupled signaling (RGS) accelerate GTP hydrolysis and consequently influence signal termination. The RGS-4 gene has recently been reported to be implicated in a wide range of neuropsychiatric disorders including schizophrenia, Alzheimer's disease and addictions. / In this study, the vervet RGS-4 gene was sequenced on a CEQ 8000 genetic analysis system (Beckman Coulter) and characterized using molecular and bioinformatic tools. The obtained vervet sequence overall showed 95.3% sequence identity with the human RGS4 gene. / Thereafter, SNPs in the region encompassing the proximal promoter, exon 1 and the first 450 bp of intron 1 were identified by direct sequencing of 8 unrelated individuals. One of the identified SNPs, +35 [A/G], was genotyped in 155 juvenile vervets previously phenotyped for personality traits, including social isolation. Although preliminary association analysis fails to attain statistical significance (p=0.074), the sample size is small. Additional genotyping of phenotypically defined individuals needs to be undertaken.

To understand the physiological role of the protein tyrosine phosphatase (PTP) PEST, the phenotypes of PTP-PEST (--/--) mouse fibroblasts have been studied PTP-PEST was shown to translocate to the membrane periphery following stimulation by the extracellular matrix protein fibronectin, and to play a role in cell migration, spreading, and cytokinesis. For this action, PTP-PEST was shown to regulate the phosphorylation levels of p130cas, paxillin, FAK and PSTPIP, and their association with SH2 domain-containing proteins like Src and Crk. From these proteins, p130cas is known to be a direct substrate. Experiments for the obtention of p130cas (--/--) embryos and cell lines have also been initiated, up to the generation of adult chimeric animals. In the process, a genomic fragment of p130cas containing a C-terminal exon has been cloned and analysed.

The TLE proteins are the mammalian homologs of Drosophila Groucho, a member of the Notch signaling pathway. Notch signaling controls the differentiation of a variety of tissues in invertebrates and vertebrates. To examine the role of the TLE genes in mammalian development, the expression of one of them, TLE 1, was analysed during mouse embryogenesis. This gene is expressed during a number of cell-determination events, including embryonic segmentation, central and peripheral neurogenesis, and epithelial differentiation. This pattern resembles groucho function in Drosophila and suggests that Groucho and TLE 1 protein may play similar developmental roles. These results also show that TLE 1 is co-expressed during a variety of cell - fate choices with several vertebrate homologs of genes implicated in the Drosophila Notch cascade, suggesting a role for the TLE 1 protein in mammalian Notch signaling.

The most common and generally best performing replacement algorithm in modern caches is LRU. Despite LRU's superiority, it is still possible that other feasible and implementable replacement policies could yield better performance. (34) found that an optimal replacement policy (OPT) would often have a miss rate 70% that of LRU. / If better replacement policies exist, they may not be obvious. One way to find better policies is to study a large number of address traces for common patterns. Such an undertaking involves such a large amount of data, that some automated method of generating and evaluating policies is required. Genetic Algorithms provide such a method, and have been used successfully on a wide variety of tasks (21). / The best replacement policy found using this approach had a mean improvement in overall hit rate of 0.6% over LRU for the benchmarks used. This corresponds to 27% of the 2.2% mean difference between LRU and OPT. Performance of the best of these replacement policies was found to be generally superior to shadow cache (33), an enhanced replacement policy similar to some of those used here.

The 5$ sp prime$ untranslated region (UTR) of c-myc and human immunodeficiency virus type 1 (HIV-1) mRNAs were used as models in a variety of in vitro and in vivo systems in order to study the role of this region in the control of eukaryotic gene expression. Using an ultraviolet light-induced crosslinking assay, a 55 kilodalton protein was identified in extracts of HeLa, mouse erythroleukemia, and other cell lines, which interacts specifically with a purine-rich RNA sequence in the 5$ sp prime$ UTR of c-myc. The function of this protein in control of c-myc expression is not known, but may be implicated in the process of transcriptional elongation. The 5$ sp prime$ UTR of HIV-1 mRNAs was shown to inhibit strongly the translation of a heterologous mRNA; this inhibition was dependent on the secondary structure predicted to form in this region, and on the accessibility of the cap structure to initiation factors. The structural requirements in the HIV-1 5$ sp prime$ UTR for trans-activation by the viral tat gene product were examined by mutagenesis studies; the base-pairing in the stem-loop structure, the sequence of the loop, and the presence of a three nucleotide bulge were found to be critical features necessary for complete trans-activation. These findings indicate that the 5$ sp prime$ UTR can have important effects on the expression of eukaryotic genes.

Atrial natriuretic factor (ANF), a 28 amino acid peptide hormone, is the major secretory product of the heart. Because of its diuretic, natriuretic and vasodilating activities, this peptide may be involved in the maintenance of proper fluid and electrolyte balance and blood pressure control. In order to study the transcriptional regulation of ANF, we have isolated the rat ANF gene and we have established a system of cardiocytes in primary cell culture for studies on the hormonal, tissue-specific and developmental regulation of the ANF gene. Using this in vitro system, as have demonstrated that thyroid hormone increases ANF mRNA levels about 2- to 4-fold in atrial and ventricular cells in primary cardiocyte cell cultures, respectively. Similarly, glucocorticoids augment by about 3-fold both atrial and ventricular ANF mRNA levels in cardiac cells in culture. Glucocorticoids exert this effect at the transcriptional level probably via the binding of glucocorticoid receptor to a DNA element in the distal 5$ sp prime$-flanking sequences of the gene as suggested by DNA-mediated transfection studies in cardiocyte cultures. In order to better understand the mechanisms governing the cardiac-specific as well as developmental expression of the ANF gene, we have analyzed ANF promoter sequences by transient transfection studies in primary cardiocyte cultures. Our data show that the ANF promoter is active only in cells of cardiac origin. Moreover, up to $-$1.6 kb of 5$ sp prime$ upstream sequences are necessary for full expression of the ANF gene in cardiac cells. Within these sequences, two particular elements, a proximal and a distal, are necessary for full ANF transcriptional activity. The proximal element can confer cardiac specificity to an otherwise non tissue-specific heterologous promoter. Further upstream sequences, between $-$2.5 and $-$1.6 kb appear to be implicated in the developmental control of ANF gene expression, as assessed by differential activity in 1 and 4 d

A method is described that allows one to assess heritability of brain structures using magnetic resonance images (MRI). We have obtained MRI images for a group of identical twins and another group of fraternal twins. Each MRI image was segmented into white and gray matter then smoothed to estimate white and gray matter density at each 3D image element or voxel. At each voxel, we use the intraclass correlations of the two groups in order to compute a 3D image of a measure of heritability, as well as a test statistic for detecting those regions with significant heritability. To our knowledge the genetic variability of white and gray matter density has not been investigated on a voxel basis previously.

Previous studies have attempted to associate aspects of course and outcome with the schizophrenia genotype. Reliance on a dichotomous familial loading variable as the single indicator of genetic loading has yielded inconsistent results. The goal of the present study was to create a continuous loading variable based on multiple indicators of the schizophrenia genotype (family history of schizophrenia and attention and verbal memory in the 1st degree relatives of the patient) and to determine its association with premorbid adjustment, age at onset, symptoms, and chronicity. Our results suggest that a continuous family history variable predicted problems in premorbid adjustment and an earlier age at onset. Hierarchical regression analysis revealed that verbal memory in the siblings of the patient increased the amount of variance accounted for in premorbid attention problems, over and above that predicted by family history. A "genetic loading" variable based on the incorporation of neuropsychologically impaired siblings into the calculations of family history was significantly correlated with premorbid attention problems.

Salmonellosis is an important foodborne disease with world-wide economic and public health significance. Domestic poultry constitute one of the largest reservoirs of Salmonella in animal populations. Susceptibility to this infection varies in different chicken lines suggesting the involvement of host genetic factors. The gene, Nramp1, and the locus, Lps, are well-characterized in the mouse and are known to be involved in the early, innate responses to Salmonella infection. In our laboratory, we have used a comparative gene approach to examine whether NRAMP1 and LPS are important to chicken salmonellosis. / This thesis has continued our comparative examination through three objectives. The first objective was to determine the frequency of the NRAMP1 Arg223→Gln223 variant, originally found in Salmonella-susceptible chicken line C, in outbred chicken and pedigreed stock. Our second objective was to genetically map chicken TNC, a closely linked marker to LPS. A chicken backcross reference population was used to map this gene: the Compton mapping panel. We mapped TNC to microchromosome E41W17 in a region of conserved synteny with mouse chromosome 4 and human chromosome 9q. The third objective of this thesis was to develop chicken NRAMP1 antibodies in order to eventually characterize the protein's function and expression. The antisera raised in New Zealand White rabbits identified a 90--100 kDa species on a Western blot that most likely corresponds to the glycosylated chicken NRAMP1.