Brucella abortus and Agrobacterium tumefaciens are pathogenic bacteria that infect animals and plants, respectively. These bacteria are genetically similar and are found within the same Class, Alphaproteobacteria, and Order, Rhizobiales, of the domain Eubacteria; however, they survive and replicate in vastly different environmental niches. In Order to adapt to different environments, bacteria utilize several mechanisms of gene regulation to tightly control gene expression. Two of these mechanisms include transcriptional regulators and small regulatory RNAs (sRNAs), which can activate and repress gene expression through various interactions with DNA, mRNA, and proteins. A well-conserved transcriptional regulator among the Rhizobiales is VtlR, a virulence-associated transcriptional LysR regulator. The objectives of this dissertation were three fold: 1) characterize the known regulon of VtlR in B. abortus with regards to gene regulatory function and virulence, 2) determine the regulon of VtlR in A. tumefaciens and define the mechanism by which this regulation occurs, and 3) define the role of an ABC-type transport system indirectly regulated by VtlR in B. abortus that putatively imports the non-proteinogenic amino acid gamma-aminobutyric acid (GABA). VtlR was characterized in B. abortus as a virulence-associated transcriptional regulator that directly activates four genes: the sRNA AbcR2, and the three small hypothetical proteins BAB1_0914, BAB2_0512, and BAB2_0574; and deletion of vtlR led to a significant defect in the ability of B. abortus to cause infection in vitro and in vivo. Since dysregulation of abcR2 alone could not account for the defect in virulence, it was hypothesized that one or all three hypothetical proteins could be responsible for a virulence phenotype observed in ΔvtlR. This turned out to not be the case, as a deletion of the entire VtlR regulon displayed no difference in virulence compared to the parental strain. Further characterization of the small hypothetical proteins is outlined in Chapter 2 and the data revealed bona fide translation of each small protein, and the deletion strain of the VtlR regulon displayed a growth defect when grown in the presence of the sugar fucose. This phenotype was subsequently observed in ΔvtlR as well. This led to the identification of a putative fucose transport and metabolism locus in B. abortus that has yet to be studied. In A. tumefaciens, VtlR is necessary for proper attachment to plant cells and biofilm formation and regulates over 200 genes, significantly more than the four genes VtlR regulates in B. abortus. The mechanism by which this occurs was unknown, and the relationship between VtlR and AbcR1 or AbcR2 was uncharacterized. The data in Chapter 3 outline the VtlR network by showing that VtlR regulation of myriad genes in A. tumefaciens is primarily indirect via the direct regulation of a few sRNAs. This direct interaction was shown experimentally and a VtlR binding box was identified in the A. tumefaciens genome. This project outlines the divergence of a regulatory element between phylogenetically related organisms that occupy different environmental niches. The AbcR sRNAs are conserved throughout the Rhizobiales and regulate numerous ABC-type transport systems within these bacteria. In A. tumefaciens, one of these transport systems specifically transports the amino acds proline and GABA. B. abortus contains homologs of this system, which led to the hypothesis that the brucellae may also transport GABA but for a yet unknown purpose. The data in Chapter 4 revealed that B. abortus also transports GABA in vitro and this transport is under the regulation of AbcR1 and AbcR2. This transport was increased under extreme nutrient limitations and was uninhibited by the presence of other amino acids. Metabolic studies showed GABA is not utilized by B. abortus under aerobic conditions, and transcriptomic data revealed increased expression of several loci in the presence of GABA. Altogether, this study uncovers a putative signaling role for the amino acid GABA that has been understudied in bacterial pathogens that infect animal hosts. Overall, the work presented in this dissertation is focused on further elucidating the biological role of downstream regulatory targets of both VtlR and the sRNAs AbcR1 and AbcR2 in the related organisms Brucella abortus and Agrobacterium tumefaciens. Findings show that while there are similarities between the two systems, there are also many differences that may be attributed to the vastly different lifestyles of each organism. / Doctor of Philosophy / Brucella abortus and Agrobacterium tumefaciens are two highly related bacterial pathogens that infect mammals and plants, respectively. Although genetically related, both organisms survive and replicate in vastly different environmental niches with one living in the soil (i.e., A. tumefaciens) and the other living within immune cells of the infected host (i.e., B. abortus). In Order to quickly adapt to changing environmental conditions, the bacteria must rapidly control gene expression through multiple regulatory mechanisms. The works presented in this dissertation will focus on further characterizing one of these regulatory systems and comparing the homologous systems shared by B. abortus and A. tumefaciens. This includes uncovering a putative sugar transport and metabolism system, as well as discovering the potential for host-pathogen signaling via the well-studied neurotransmitter GABA.

<p>Data science uses the latest techniques in statistics and machine learning to extract insights from data. With the increasing amount of protein data, a number of novel research approaches have become feasible.</p><p>Micropeptides are an emerging field in the protein universe. They are small proteins with <= 100 amino acid residues (aa) and are translated from small open reading frames (sORFs) of <= 303 base pairs (bp). Traditionally, their existence was ignored because of the technical difficulties in isolating them. With technological advances, a growing number of micropeptides have been characterized and shown to play vital roles in many biological processes. Yet, we lack bioinformatics methods for predicting them directly from DNA sequences, which could substantially facilitate research in this field with minimal cost. With the increasing amount of data, developing new methods to address this need becomes possible. We therefore developed MiPepid, a machine-learning-based method specifically designed for predicting micropeptides from DNA sequences by curating a high-quality dataset and by training MiPepid using logistic regression with 4-mer features. MiPepid performed exceptionally well on holdout test sets and performed much better than existing methods. MiPepid is available for downloading, easy to use, and runs sufficiently fast.</p><p>Long noncoding RNAs (LncRNAs) are transcripts of > 200 bp and does not encode a protein. Contrary to their “noncoding” definition, an increasing number of lncRNAs have been found to be translated into functional micropeptides. Therefore, whether most lncRNAs are translated is an open question of great significance. To address this question, by harnessing the availability of large-scale human variation data, we have explored the relationships between lncRNAs, micropeptides, and canonical regular proteins (> 100 aa) from the perspective of genetic variation, which has long been used to study natural selection to infer functional relevance. Through rigorous statistical analyses, we find that lncRNAs share a similar genetic variation profile with proteins regarding single nucleotide polymorphism (SNP) density, SNP spectrum, enrichment of rare SNPs, etc., suggesting lncRNAs are under similar negative selection strength with proteins. Our study revealed similarities between micropeptides, lncRNAs, and canonical proteins and is the first attempt to explore the relationships between the three groups from a genetic variation perspective.</p><p>Deep learning has been tremendously successful in 2D image recognition. Protein binding ligand prediction is fundamental topic in protein research as most proteins bind ligands to function. Proteins are 3D structures and can be considered as 3D images. Prediction of binding ligands of proteins can then be converted to a 3D image classification problem. In addition, a large number of protein structure data are available now. We therefore utilized deep learning to predict protein binding ligands by designing a 3D convolutional neural network from scratch and by building a large 3D image dataset of protein structures. The trained model achieved an average F1 score of over 0.8 across 151 classes on the holdout test set. Compared to existing methods, our model performed better. In summary, we showed the feasibility of deploying deep learning in protein structure research.</p><p>In conclusion, by exploring various edges of the protein universe from the perspective of data science, we showed that the increasing amount of data and the advancement of data science methods made it possible to address a wide variety of pressing biological questions. We showed that for a successful data science study, the three components – goal, data, method – all of them are indispensable. We provided three successful data science studies: the careful data cleaning and selection of machine learning algorithm lead to the development of MiPepid that fits the urgent need of a micropeptide prediction method; identifying the question and exploring it from a different angle lead to the key insight that lncRNAs resemble micropeptides; applying deep learning to protein structure data lead to a new approach to the long-standing question of protein-ligand binding. The three studies serve as excellent examples in solving a wide range of data science problems with a variety of issues.</p>