This paper presents a novel MEMS based DNA computer for solving SAT problems. No time-consuming sample preparation procedures and delicate sample applying equipment were required for the computing process. Moreover, experimental results show the bound DNA sequences can sustain the chemical solutions during computing processes such that the proposed method shall be useful in dealing with large scale problems. An algorithm based on a modified sticker model accompanied with a state-of-the-art MEMS-based microarray experiment is demonstrated to solve SAT problem which has long served as a benchmark in DNA computing. Unlike conventional DNA computing algorithms need an initial data pool to cover all correct and incorrect answers and further execute a series of separation procedures to destroy the unwanted ones, we built solutions in parts to satisfy one clause in one step, and eventually solve the entire Boolean formula through steps. Accordingly this algorithm greatly reduces the formation of unnecessary candidate solutions and shall be very practical as problem size grows. In this study, a novel MEMS-based technology including utilizing blank mask as the microarray substrate to prevent the self-fluorescent effect, a twin-mask back-side exposure process to improve the computing speed and a low-temperature backing process to prevent DNA damage during computing procedure. In addition, the minimal time requirement for DNA hybridization was also evaluated experimentally. The paper reports a novel computing method for solving SAT problem utilizing a state-of-art MEMS-based microarray. The advantage of this method is as the problem size scales up, it only needs to linearly increase the variety of sequences standing for variables and augment the array size. Therefore, while solving a complicated SAT problem, the numbers of DNA sample and the time for the computing process can be dramatically reduced with this approach.

Global gene expression in Sorghum bicolor, an important crop showing drought tolerance in arid and semi-arid cultivated areas, was monitored to exposure of 8-days seedlings to water deficit (20% polyethylene glycol) or cold stress (4 ºC). A sorghum cDNA microarray, including ~13,000 (milestone version 1) or ~28,000 (milestone version 2) unigenes, was used to examine gene expression in shoots and roots at 3 and 27hours after stress treatment. ~1,300 and ~2,300 genes were modulated by water deficit and cold stress, respectively. Up-regulated genes included previously identified stressinduced genes such as early drought-induced gene, dehydrin, late embryogenesis abundant gene, glycin and proline-rich gene, and water stress-inducible genes as well as unknown genes. Genes involved in signal transduction, lipid metabolism, transporter, and carbohydrate metabolism are induced. Quantitative real-time PCR was used to quantify changes in relative mRNA abundance for 333 and 108 genes in response to water deficit and cold stress, respectively. Stress-induced genes were classified by kinetics. Eighteen of 108 cold-induced genes were modulated by cold but not by ABA and PEG treatment. This research provides the starting point for detailed analysis and comparison of water deficit and cold modulated gene networks in sorghum.

Cardiovascular disease (CVD) is the leading cause of death in the United States. Oxidative stress plays an important role in the pathogenesis of CVD. Heart failure is the end point of many forms of CVD. The purpose of this study is to identify novel cardiac specific indicators of oxidative injury useful for early and convenient diagnosis of heart failure.To determine the most suitable method for identification of non-invasive oxidative injury indicators in general, human diploid fibroblasts (HDFs) were treated with H2O2 for collection of mRNA, cell lysates and conditioned media to perform cDNA microarray and LC-MS/MS based Multidimensional Protein Identification Technology (MudPIT) analyses. Electron Spray Ionization (ESI)-LC-MS/MS analysis of the conditioned media led to the finding of IGFBP-6 as a non-invasive biomarker of cell oxidative injuy in vitro and in vivo. The data obtained from this study indicate that proteomic analysis of conditioned media is useful to identify non-invasive biomarkers valuable for diagnosis or management of diseases.Cardiomyocyts (CMCs) and Cardiac fibroblasts (CFs) in culture were used to identify cardiac specific indicators of oxidative stress. Increased level of Cystatin C was detected in the conditioned medium of CMCs due to H2O2 treatment. In vivo models of oxidative stress were used to validate the increase of Cystatin C. Cystatin C levels increased in the plasma of mice with doxorubicin induced cardiomyopathy and coronary artery occlusion induced myocardial infarction (MI). These data indicate that Cystatin C can be a potential indicator of CMC oxidative injury in vitro and in vivo.Cystatin C is a cysteine protease inhibitor. The finding that oxidative stress induces Cystatin C led us to investigate a novel pathway regulating cardiac extracellular matrix (ECM) with CFs in culture, increased levels of ECM protein and decreased levels of Cathepsin B (CTB) protein and activity were detected upon Cystatin C treatment. With coronary artery occlusion induced MI mouse model, increased levels of Cystatin C and ECM protein and decreased levels of CTB protein and activity were detected in the infarcted area of the myocardium. These data indicate that Cystatin C serves as a potential fibrotic factor during myocardial remodeling.

A principal concern in crop agriculture is yield, and a key factor for crop growth is the availability of nitrogen. The large amount of nitrogen fertilizer required by plants is a major cost to farmers. Moreover, environmental issues such as groundwater pollution arise from the utilization of nitrogen fertilizers. Therefore, improvement in the nitrogen use efficiency (NUE) of plants is of urgent importance for sustainable and efficient agriculture. Although hybrid varieties have increased crop yields in low N conditions, the molecular mechanism of plant adaptation to N stress is not completely understood. Herein, the study of responses to N limitations in the natural signalling pathways of model plants facilitates the understanding of complex responses in plants to N stress, and this information can be used to further improve NUE. In this research, the transcriptomes of three model plants Arabidopsis, maize, and rice were compared under diverse N growth conditions. An evaluation of the response of the three plants to varying N levels was also conducted. From a statistical point of view, three distinct methods of detecting differential expression were utilized to reduce the likelihood of false positives due to the tens of thousands of genes simultaneously studied. Furthermore, the performance of three statistical approaches was compared during detection of the N-responsive genes. Finally, a clustering analysis (agglomerative hierarchical clustering) was performed on the genes that significantly responded to N levels as identified by a more biologically intuitive method called Rank Products (RP).

The field of Bioinformatics and Computational Biology (BCB) is a multidisciplinary field that has emerged due to the computational demands of current state-of-the-art biotechnology. BCB deals with the storage, organization, retrieval, and analysis of biological datasets, which have grown in size and complexity in recent years especially after the completion of the human genome project. The advent of Microarray technology in the 1990s has resulted in the new concept of high throughput experiment, which is a biotechnology that measures the gene expression profiles of thousands of genes simultaneously. As such, Microarray requires high computational power to extract the biological relevance from its high dimensional data. Current general purpose processors (GPPs) has been unable to keep-up with the increasing computational demands of Microarrays and reached a limit in terms of clock speed. Consequently, Field Programmable Gate Arrays (FPGAs) have been proposed as a low power viable solution to overcome the computational limitations of GPPs and other methods. The research presented in this thesis harnesses current state-of-the-art FPGAs and tools to accelerate some of the most widely used data mining methods used for the analysis of Microarray data in an effort to investigate the viability of the technology as an efficient, low power, and economic solution for the analysis of Microarray data. Three widely used methods have been selected for the FPGA implementations: one is the un-supervised Kmeans clustering algorithm, while the other two are supervised classification methods, namely, the K-Nearest Neighbour (K-NN) and Support Vector Machines (SVM). These methods are thought to benefit from parallel implementation. This thesis presents detailed designs and implementations of these three BCB applications on FPGA captured in Verilog HDL, whose performance are compared with equivalent implementations running on GPPs. In addition to acceleration, the benefits of current dynamic partial reconfiguration (DPR) capability of modern Xilinx’ FPGAs are investigated with reference to the aforementioned data mining methods. Implementing K-means clustering on FPGA using non-DPR design flow has outperformed equivalent implementations in GPP and GPU in terms of speed-up by two orders and one order of magnitude, respectively; while being eight times more power efficient than GPP and four times more than a GPU implementation. As for the energy efficiency, the FPGA implementation was 615 times more energy efficient than GPPs, and 31 times more than GPUs. Over and above, the FPGA implementation outperformed the GPP and GPU implementations in terms of speed-up as the dimensionality of the Microarray data increases. Additionally, the DPR implementations of the K-means clustering have shown speed-up in partial reconfiguration time of ~5x and 17x over full chip reconfiguration for single-core and eight-core implementations, respectively. Two architectures of the K-NN classifier have been implemented on FPGA, namely, A1 and A2. The K-NN implementation based on A1 architecture achieved a speed-up of ~76x over an equivalent GPP implementation whereas the A2 architecture achieved ~68x speedup. Furthermore, the FPGA implementation outperformed the equivalent GPP implementation when the dimensionality of data was increased. In addition, The DPR implementations of the K-NN classifier have achieved speed-ups in reconfiguration time between ~4x to 10x over full chip reconfiguration when reconfiguring portion of the classifier or the complete classifier. Similar to K-NN, two architectures of the SVM classifier were implemented on FPGA whereby the former outperformed an equivalent GPP implementation by ~61x and the latter by ~49x. As for the DPR implementation of the SVM classifier, it has shown a speed-up of ~8x in reconfiguration time when reconfiguring the complete core or when exchanging it with a K-NN core forming a multi-classifier. The aforementioned implementations clearly show FPGAs to be an efficacious, efficient and economic solution for bioinformatics Microarrays data analysis.

The advent of cDNA microarray technology has allowed the study of diseases such as epithelial ovarian cancer (EOC) to occur at an unprecedented level of molecular resolution. (For complete abstract open document)

In 2003, Pomeroy et al. published a research study that described a gene expression based prediction of central nervous system embryonal tumour (CNS) outcome. Over a half of decade, many models and approaches have been developed based on experimental data consisting of 99 samples with 7,129 genes. The way, how meaningful knowledge from these models can be extracted, and how this knowledge for further research is still a hot topic. This thesis addresses this and has developed an information method that includes modelling of interactive patterns, important genes discovery and visualisation of the obtained knowledge. The major goal of this thesis is to discover important genes responsible for CNS tumour and import these genes into a well structured knowledge framework system, called Brain-Gene-Ontology. In this thesis, we take the first step towards finding the most accurate model for analysing the CNS tumour by offering a comparative study of global, local and personalised modelling. Five traditional modelling approaches and a new personalised method – WWKNN (weighted distance, weighted variables K-nearest neighbours) – are investigated. To increase the classification accuracy and one-vs.-all based signal to- noise ratio is also developed for pre-processing experimental data. For the knowledge discovery, CNS-based ontology system is developed. Through ontology analysis, 21 discriminate genes are found to be relevant for different CNS tumour classes, medulloblastoma tumour subclass and medulloblastoma treatment outcome. All the findings in this thesis contribute for expanding the information space of the BGO framework.

HOX11 is a homeodomain transcription factor that has been implicated in leukaemic transformation associated with T-cell acute lymphoblastic leukaemia (T-ALL). Its role in leukaemogenesis remains enigmatic, nevertheless, in vitro and in vivo studies have provided additional evidence supporting the role of HOX11 as an oncogene. The mechanism by which HOX11 transforms cells is yet to be elucidated, however, HOX11 has been postulated to function by binding regulatory elements within the promoter regions of specific target genes in order to control gene transcription. The identification of transcriptional targets is thus thought to be critical to our understanding of the pathways controlled by this master gene regulator. To date, only three candidate HOX11 target genes have been reported and given that HOX11 overexpression can have a profound impact on cell behaviour, it is likely that many more exist. In this study, we sought to further understand the role of HOX11 in tumorigenesis by: 1) The identification of novel putative HOX11 target genes by profiling gene expression in response to HOX11 in a number of cell lines using a combination of RDA, cDNA microarray and GeneChip approaches and 2) confirming target gene status by assessing whether the proximal promoters of the leading candidates identified are transcriptionally regulated by HOX11. To identify genes whose expression was altered by HOX11, three techniques were employed, namely representational difference analysis, cDNA microarray and Affymetrix GeneChip array. Because of the relative novelty of these technologies, all three methods were employed in a complementary manner. While representational difference analysis did not require dedicated equipment and enabled the identification of novel genes, the technique was labour-intensive and also exhibited a number of problems including high levels of background. Emphasis was therefore placed on the more systematic microarray approaches that enabled a global investigation of expression patterns and thus the identification of a range of candidate target genes. Initially, this involved cDNA microarray experiments, however, during the course of this work Affymetrix GeneChip technology became available. The latter was identified as the most appropriate technology for the identification of candidate target genes because of its relative ease of use, as well as its employment of multiple independent probe pairs which greatly improved background noise, increased the range and accuracy of detection, minimized the effects of cross hybridization and drastically reduced the rate of false positives and miscalls. Using these combined approaches, several genes of interest were identified which were differentially regulated in the presence of HOX11 and thus may represent oncogenically or physiologically relevant target genes. These included OSTEOPONTIN, PAG, GUANOSINE DIPHOSPHATE DISSOCIATION INHIBITOR 3, SUR8, GAS3, C-KIT, VEGFC, NOR1 and SMARCD3. In order to confirm their role as target genes, four candidates (C-KIT, VEGFC, NOR1 and SMARCD3) were characterized in terms of the ability of their proximal promoters to be transcriptionally regulated by HOX11 using luciferase reporter assays. Significant repression of the proximal promoters of C-KIT and VEGFC by HOX11 was observed, which provided further evidence for their status as target genes. This repression was, however, in stark contrast to the transcriptional activation seen when the C-KIT and VEGFC proximal promoters were co-transfected with a HOX11 mutant lacking the third helix of the DNA-binding homeodomain. This unexpected finding suggested that the transcriptional activity of HOX11 is complex and highly context-dependent, and in particular, highlighted the importance of an intact homeodomain for HOX11 function. C-KIT and VEGFC are both involved in tyrosine kinase signal transduction pathways, as a receptor tyrosine kinase and tyrosine kinase ligand, respectively. C-KIT plays an important role in the survival and self-renewal of haematopoietic cells. It is a previously identified and relatively well characterized oncogene known to be regulated by other transcription factors (SCL/TAL1 and LMO) implicated in the pathogenesis of T-ALL. VEGFC is a member of the vascular endothelial growth factor family that functions in angiogenesis and lymphangiogenesis. A paracrine loop involving VEGFC and its receptor VEGFR-3 has previously been implicated in leukaemic cell survival. While further work is required in order to confirm the status of VEGFC and C-KIT as oncogenically-relevant HOX11 target genes and to characterize their exact mode of regulation, these findings implicate receptor tyrosine kinases in HOX11-mediated tumorigenesis and underscore their potential importance as therapeutic targets in haematological malignancies.

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

München, Techn. University, Diss., 2007.