The PCR-based genome walking method has been commonly used to isolate upstream regions from known cDNA sequences. The limitation of this technique is based on the location of the restriction site upstream to the gene-specific primer in the genome; hence, different restriction enzymes have to be used to isolate larger upstream fragments. In this paper, we present the advantageous use of partial and size-selected DNA as templates for genome walking, in isolating larger upstream fragments. We have successfully tested this approach to isolate larger upstream fragments using the FailSafe™ PCR System. Use of partial digestion and size selection can provide better chances in obtaining larger flanking regions of known DNA sequence, when compared to use of total digested DNA.

Aedes aegypti (Ae. aegypti) is one of the most important vectors of dengue, chikungunya and yellow fever viruses. The use of chemical control strategies such as insecticides is associated with problems including the development of insecticide resistance, side effects on animal and human health, and environmental concerns. Because current methods have not proven sufficient to control these diseases, developing novel, genetics-based, control strategies to limit the transmission of disease is urgently needed. Increased knowledge about mosquito-pathogen relationships and the molecular biology of mosquitoes now makes it possible to generate transgenic mosquito strains that are unable to transmit various parasites or viruses. Ae. aegypti genetic experiments are enabled, and limited by, the catalog of promoter elements available to drive transgene expression. To find a promoter able to drive robust expression of firefly (FF) luciferase in Ae. aegypti embryos, an experiment was designed to compare Ae. aegypti endogenous and exogenous promoters. The PUb promoter was found to be extremely robust in expression of FF luciferase in different stages of embryonic development from 2-72 hours after injection. In subsequent experiments, transformation frequency was calculated using four different promoters (IE1, UbL40, hsp82 and PUb) to express the Mos1 transposase open reading frame in Mos1-mediated transgenesis. Germline transformation efficiency and size of transgenic cluster were not significantly different when using endogenous Ae. aegypti PUb or the commonly used exogenous Drosophila hsp82 promoter to express Mos1 transposase. This study also describes the development of new tools for gene editing in the Ae. aegypti mosquito genome and the use of these tools to design an efficient gene drive system in this mosquito. Homing endonucleases (HEs) are selfish elements which catalyze double-stranded DNA (dsDNA) breaks in a sequence-specific manner. The activities of four HEs (Y2-I-AniI, I-CreI, I-PpoI, and I-SceI) were investigated for their ability to catalyze the excision of genomic segments from the Ae. aegypti genome. All four enzymes were found to be active in Ae. aegypti; however, the activity of Y2-I-AniI was higher compared to the other three enzymes. Single-strand annealing (SSA) and non-homologous end-joining (NHEJ) pathways were identified as mechanisms to repair HE-induced dsDNA breaks. TALE nucleases (TALENs) are a group of artificial enzymes capable of generating site-specific DNA lesions. To examine the ability of TALENs for gene editing in Ae. aegypti, a pair of TALENs targeted to the kmo gene were expressed from a plasmid following embryonic injection. Twenty to forty percent of fertile G0 produced white-eyed progeny which resulted from disruption of the kmo gene. Most of these individuals produced more than 20% white-eyed progeny, with some producing up to 75%. A small deletion of one to seven bp occurred at the TALEN recognition site. These results show that TALEN and HEs are highly active in the Ae. aegypti germline and can be used for gene editing and gene drive strategies in Ae. aegypti. / Ph. D.

Promoter region of protein-coding genes are gradually being well understood, yet no comparable studies exist for the promoter of long non-coding RNA (lncRNA) genes which has emerged as a global potential regulator in multiple cellular process and different diseases for human. To understand the difference in the transcriptional regulation pattern of these genes, previously, we proposed a machine learning based model to classify the promoter of protein-coding genes and lncRNA genes. In this study, we are presenting DeepCNPP (deep coding non-coding promoter predictor), an improved model based on deep learning (DL) framework to classify the promoter of lncRNA genes and protein-coding genes. We used convolution neural network (CNN) based deep network to classify the promoter of these two broad categories of human genes. Our computational model, built upon the sequence information only, was able to classify these two groups of promoters from human at a rate of 83.34% accuracy and outperformed the existing model. Further analysis and interpretation of the output from DeepCNPP architecture will enable us to understand the difference in transcription regulatory pattern for these two groups of genes.

Human immunodeficiency virus-1 (HIV-1) establishes a persistent proviral reservoir by integrating into the genome of infected host cells. Current antiretroviral treatments (ART) do not target this persistent population of proviruses which include latently infected cells that upon treatment interruption can be reactivated to contribute to HIV-1 rebound. Deep sequencing of persistent HIV-1 proviruses has revealed that greater than 90% of integrated HIV-1 genomes are defective and unable to produce infectious virions. We hypothesized that intragenic elements in the HIV genome support transcription of aberrant HIV-1 RNAs from defective proviruses that lack long terminal repeats (LTRs). Using an intact provirus detection assay, I observed that resting CD4+ T cells and monocyte-derived macrophages (MDMs) are biased towards generating defective HIV-1 proviruses. Multiplex reverse transcription droplet digital polymerase chain reaction (RT-ddPCR) identified env and nef transcripts which lacked 5’ untranslated regions (UTR) in acutely infected CD4+ T cells and MDMs indicating transcripts are generated that do not utilize the promoter within the LTR. 5’UTR-deficient env transcripts were also identified in a cohort of people living with HIV-1 (PLWH) on ART, suggesting that these aberrant RNAs are produced in vivo. Using 5’ rapid amplification of cDNA ends (RACE), I mapped the start site of these transcripts within the Env gene. This region bound several cellular transcription factors and functioned as a transcriptional regulatory element that could support transcription and translation of downstream HIV-1 RNAs. Transient expression of an HIV-1 5’UTR deletion construct in HEK293T cells demonstrated that HIV-1 transcripts and proteins are still produced when the 5’UTR is absent. These studies provide mechanistic insights into how defective HIV-1 proviruses are persistently expressed to potentially drive inflammation in PLWH.

No description available.

No description available.

No description available.

In this project, a neural network based system is used to classify the promoter regions found in Escherichia coli DNA sequences. An unsupervised algorithm based on the self-organizing feature map is used to classify the sequences and a dynamic programming algorithm is used too query the trained neural networks. In order to generalize the neural network's weights for display purposes, a back propagation supervised learning algorithm based on the conjugate gradient method is used to map the weights to one of the fifteen combinations of adenine, cytosine, guanine, and thymine (the chemical components of DNA). The results show that this method is able to classify the training sequences into discrete sub-classes which provide a query base for classifying new sequences. This method can be used for any class of sequences and can be extended for use in searching sequence databases. / Thesis / Master of Science (MS)

Researchers of <i>Clostridium perfringens</i>, a Gram-positive anaerobic pathogen, were lacking a tightlyregulated, inducible promoter system in their genetic toolbox. We constructed a lactose-inducible plasmid-based system utilizing the transcriptional regulator, BgaR. Using the <i>E. coli</i> reporter GusA, we characterized its induction in three different strains of <i>C. perfringens</i>. We then used a newly-developed mutation system to create in-frame deletion mutants in three genes with homology to Type IV pilins, and we used the promoter system described above to complement the mutants. We analyzed each pilin for localization and expression, as well as tested each of the mutants for various phenotypes frequently associated with type IV pili (TFP) and type II secretion systems. PilA2, PilA3, and PilA4 localized to the poles of the cells. PilA2 was expressed in the wildtype when <i>C. perfringens</i> was grown on agar plates, and the PilA3 mutant lacked a von Willebrand factor A domain-containing protein in its secretome. We used our promoter system to express GFP-tagged versions of the TFP ATPase homologues and view them in cells growing on surfaces. We saw that PilB1 and PilB2 co-localized nearly all of the time, while a portion of PilT was independent of the PilB proteins. PilT appeared necessary for the localization of PilB, and it localized independently of TFP proteins in <i>Bacillus subtilis</i>. PilT's typical localization in <i>Bacillus subtilis</i> was disrupted when the GTPase and polymerization activity of cell division protein FtsZ was blocked, suggesting that PilT associates with cell division proteins. / Master of Science

Combinatorial control of transcription is a well established phenomenon in the cell. Multiple transcription factors often bind to the same transcriptional control region of a gene and interact with each other to control the expression of the gene. It is thus necessary to consider the joint conservation of sequence pairs in order to identify combinations of binding sites to which the transcription factors bind. Conventional motif finding algorithms fail to address this issue. We propose a novel biclustering algorithm based on random sampling to identify candidate binding site combinations. We establish bounds on the various parameters to the algorithm and study the conditions under which the algorithm is guaranteed to identify candidate binding sites. We analyzed a yeast cell cycle gene expression data set using our algorithm and recovered certain novel combinations of binding sites, besides those already reported in the literature. / Master of Science