Single nucleotide polymorphisms (SNPs) are DNA sequence variations, occurring when a single nucleotide –A, T, C or G – is altered. Arguably, SNPs account for more than 90% of human genetic variation. Dr. Tebbutt's laboratory has developed a highly redundant SNP genotyping assay consisting of multiple probes with signals from multiple channels for a single SNP, based on arrayed primer extension (APEX). The strength of this platform is its unique redundancy having multiple probes for a single SNP. Using this microarray platform, we have developed fully-automated genotype calling algorithms based on linear models for individual probe signals and using dynamic variable selection at the prediction level. The algorithms combine separate analyses based on the multiple probe sets to give a final confidence score for each candidate genotypes. Our proposed classification model achieved an accuracy level of >99.4% with 100% call rate for the SNP genotype data which is comparable with existing genotyping technologies. We discussed the appropriateness of the proposed model related to other existing high-throughput genotype calling algorithms. In this thesis we have explored three new ideas for classification with high dimensional data: (1) ensembles of various sets of predictors with built-in dynamic property; (2) robust classification at the prediction level; and (3) a proper confidence measure for dealing with failed predictor(s). We found that a mixture model for classification provides robustness against outlying values of the explanatory variables. Furthermore, the algorithm chooses among different sets of explanatory variables in a dynamic way, prediction by prediction. We analyzed several data sets, including real and simulated samples to illustrate these features. Our model-based genotype calling algorithm captures the redundancy in the system considering all the underlying probe features of a particular SNP, automatically down-weighting any ‘bad data’ corresponding to image artifacts on the microarray slide or failure of a specific chemistry. Though motivated by this genotyping application, the proposed methodology would apply to other classification problems where the explanatory variables fall naturally into groups or outliers in the explanatory variables require variable selection at the prediction stage for robustness.

Single nucleotide polymorphisms (SNPs) are DNA sequence variations, occurring when a single nucleotide –A, T, C or G – is altered. Arguably, SNPs account for more than 90% of human genetic variation. Dr. Tebbutt's laboratory has developed a highly redundant SNP genotyping assay consisting of multiple probes with signals from multiple channels for a single SNP, based on arrayed primer extension (APEX). The strength of this platform is its unique redundancy having multiple probes for a single SNP. Using this microarray platform, we have developed fully-automated genotype calling algorithms based on linear models for individual probe signals and using dynamic variable selection at the prediction level. The algorithms combine separate analyses based on the multiple probe sets to give a final confidence score for each candidate genotypes. Our proposed classification model achieved an accuracy level of >99.4% with 100% call rate for the SNP genotype data which is comparable with existing genotyping technologies. We discussed the appropriateness of the proposed model related to other existing high-throughput genotype calling algorithms. In this thesis we have explored three new ideas for classification with high dimensional data: (1) ensembles of various sets of predictors with built-in dynamic property; (2) robust classification at the prediction level; and (3) a proper confidence measure for dealing with failed predictor(s). We found that a mixture model for classification provides robustness against outlying values of the explanatory variables. Furthermore, the algorithm chooses among different sets of explanatory variables in a dynamic way, prediction by prediction. We analyzed several data sets, including real and simulated samples to illustrate these features. Our model-based genotype calling algorithm captures the redundancy in the system considering all the underlying probe features of a particular SNP, automatically down-weighting any ‘bad data’ corresponding to image artifacts on the microarray slide or failure of a specific chemistry. Though motivated by this genotyping application, the proposed methodology would apply to other classification problems where the explanatory variables fall naturally into groups or outliers in the explanatory variables require variable selection at the prediction stage for robustness.

Thesis (Ph.D.)--Virginia Commonwealth University, 2008. / Title from title-page of electronic thesis. Prepared for: Dept. of Biostatistics. Bibliography: leaves 215-233.

Thesis (Ph.D.)--Virginia Commonwealth University, 2008. / Prepared for: Dept. of Biostatistics. Bibliography: leaves 215-233. Also available online via the Internet.

Thesis (Ph.D.)--Virginia Commonwealth University, 2010. / Prepared for: Dept. of Biostatistics. Title from title-page of electronic thesis. Bibliography: leaves 166-187.

Single nucleotide polymorphisms (SNPs) are DNA sequence variations, occurring when a single nucleotide –A, T, C or G – is altered. Arguably, SNPs account for more than 90% of human genetic variation. Dr. Tebbutt's laboratory has developed a highly redundant SNP genotyping assay consisting of multiple probes with signals from multiple channels for a single SNP, based on arrayed primer extension (APEX). The strength of this platform is its unique redundancy having multiple probes for a single SNP. Using this microarray platform, we have developed fully-automated genotype calling algorithms based on linear models for individual probe signals and using dynamic variable selection at the prediction level. The algorithms combine separate analyses based on the multiple probe sets to give a final confidence score for each candidate genotypes. Our proposed classification model achieved an accuracy level of >99.4% with 100% call rate for the SNP genotype data which is comparable with existing genotyping technologies. We discussed the appropriateness of the proposed model related to other existing high-throughput genotype calling algorithms. In this thesis we have explored three new ideas for classification with high dimensional data: (1) ensembles of various sets of predictors with built-in dynamic property; (2) robust classification at the prediction level; and (3) a proper confidence measure for dealing with failed predictor(s). We found that a mixture model for classification provides robustness against outlying values of the explanatory variables. Furthermore, the algorithm chooses among different sets of explanatory variables in a dynamic way, prediction by prediction. We analyzed several data sets, including real and simulated samples to illustrate these features. Our model-based genotype calling algorithm captures the redundancy in the system considering all the underlying probe features of a particular SNP, automatically down-weighting any ‘bad data’ corresponding to image artifacts on the microarray slide or failure of a specific chemistry. Though motivated by this genotyping application, the proposed methodology would apply to other classification problems where the explanatory variables fall naturally into groups or outliers in the explanatory variables require variable selection at the prediction stage for robustness. / Science, Faculty of / Statistics, Department of / Graduate

Diss. (sammanfattning) Uppsala : Uppsala universitet, 2005. / Härtill 5 uppsatser.

RNA editing by adenosine deamination is catalyzed by members of an enzyme family known as adenosine deaminases that act on RNA (ADARs). ADARs can change the structure of RNA by changing an AU base-pair to an IU mismatch. This frequently modifies the function of the encoded protein, and an emerging theme associated with A-to-I mRNA editing is that tissues often regulate the ratio of proteins expressed from edited and unedited mRNAs to fine-tune cellular responses and functions. In mammals, pre-mRNA of receptor proteins involved in neurotransmission, including serotonin receptors and glutamate receptors, are edited. Currently, only a limited number of human ADAR substrates are known, whereas indirect evidence suggests a substantial fraction of all pre-mRNAs being affected. To identify RNAs containing inosine residues, this study used a multi step approach; including (1) inosine-specific base cleavage and RNase T1 digestion, (2) purification of polyA-tailed mRNA, (3) RT w/ T7-polydT primer, (4) probe synthesis and microarray analysis. Using this method it is possible to identify novel targets of A to I editing. Approximately 100 genes showed a significant decrease in two arrays. Future analysis of these targets should reveal the biomedical significance of A-to-I editing.

Diss. (sammanfattning) Umeå : Univ., 2007. / Härtill 6 uppsatser.

The parasitic angiosperm Orobanche aegyptiaca (Pers.) (Egyptian broomrape) is a root holoparasite that causes severe losses in yield and quality of many crops. Control of Orobanche is extremely challenging, in part because the parasite is hidden underground for most of its life cycle. However, the dependence of the parasite on the host suggests that broomrape-resistant hosts could be an ideal control method. Genetic engineering strategies may facilitate realization of this goal, but require an understanding of host defense responses to parasitism. Previous studies with tobacco indicated that broomrape parasitism induces host genes associated with jasmonic acid (JA)-mediated defenses such as wound responses and localized production of phenylpropanoid and isoprenoid phytoalexins. However, the gene for the pathogenesis-related (PR) protein, PR-1a, was not induced by parasitism in tobacco. Expression of PR-1a is correlated with the salicylic acid (SA)-mediated defense pathway that leads to systemic acquired resistance (SAR). The objective of this research was to extend the characterization of PR gene expression in order to define the scope of host defense response. Analyses of gene expression using RNA hybridization and RT-PCR in broomrape-parasitized Arabidopsis thaliana roots indicated that PR-1, PR-2, PR-5, as well as the JA-associated PDF1.2, were slightly induced by parasitism. Expression of PR-1, PR-5, and PDF1.2 in parasitized roots was not detectable by RNA hybridization analysis, but was demonstrated by RT-PCR. Interestingly, shoots of the parasitized plants showed greater PR gene expression levels than roots, indicating that O. aegyptiaca induced a response in the host that was systemic and amplified in shoots. Microarray analysis of parasitized Arabidopsis roots demonstrated a broad range of host gene expression changes including both defense- and non-defense-related genes. Genes induced were consistent with O. aegyptiaca preferentially stimulating JA-mediated responses. The failure of O. aegyptiaca to elicit SA-mediated defenses in host roots suggested that exogenous induction of this signaling pathway could enhance host resistance to parasitism. Treatment of O. aegyptiaca-inoculated tobacco with BTH, a SA analog that activates SAR, caused a 49% reduction in O. aegyptiaca numbers. Analysis of PR-1a using RNA hybridizations and protein immunoblots in treated plants showed the expected induction in shoots, but not in roots, confirming the organ-specific differences in defense response observed in Arabidopsis. Experiments using a strategy to engineer the hypersensitive response via the gene-for-gene interaction confirmed previous findings that parasite-specific activation of an R/Avr interaction in tobacco reduced parasitism by approximately 50%. This research suggests that approaches to stimulate SAR in susceptible host plants may be useful for reducing Orobanche parasitism / Master of Science