Bacteriophage T7 RNA polymerase (T7 RNAP) is frequently used for RNA synthesis with unnatural base pairs (UBPs) and specific synthetic alterations in a wide range of biotechnological and medicinal applications. As a knowledge gap in the area, however, the molecular basis for recognition and processing of UBPs by T7 RNAP during transcription remains poorly understood. We explored how the hydrophobic Ds–Pa pair is recognized and processed as a third base pair by T7 RNAP during transcription elongation. T7 RNAP integrates DsTP opposite Pa with great efficiency, equivalent to that of natural nucleotides, although the kinetics of PaTP incorporation opposite Ds is significantly slower. Using structural biology approach, we discovered that T7 RNAP recognizes unnatural substrates differently than its natural substrates. We found distinct unnatural nucleoside triphosphate binding sites for PaTP and DsTP at the pre-insertion state. We identified several separate-of-function mutants of T7 RNAP that affect UBP transcription selectively but not normal nucleic acid transcription. These results provide molecular insights into the recognition of UBP transcription by T7 RNAP. Our investigations also provide important information for the creation of the next generation of UBPs for efficient transcription and other applications.

RNA polymerase II (pol II) recognizes many obstacles during transcription elongation, including DNA damage lesions and modifications, via specific interactions and leads to distinct transcriptional outcomes. We investigate three specific types of modifications/lesions in DNA and how they affect the pol II transcription process: 1) unnatural synthetic nucleotides (dNaM and dTPT3), 2) regioisomeric alkylated thymidine lesions (O2-, N3-, O4-EtdT), and 3) non-covalent minor groove DNA binders pyrrole-imidazole (Py-Im) polyamides. In Chapter 1, we investigate pol II transcription and elongation in the presence of synthetic nucleotides (dNaM and dTPT3), and the ability of pol II to distinguish between natural NTPs and the unnatural triphosphates. Selective incorporation of rNaM by pol II only occurs when dTPT3 is in the template strand, and loses its selectivity when dNaM is in the template. In Chapter 2, we discovered distinct patterns of pol II transcriptional bypass for each of the alkylated thymidine lesions. We found that pol II bypass of O2-EtdT is essentially error free, bypass of O4-EtdT is efficient and highly error prone, and bypass of N3-EtdT is extremely slow. In Chapter 3, we found that Py-Im polyamides bound to the minor groove at sequence specific sites causes prolonged pol II arrest upstream of the binding site, due to two specific residues in the pol II Switch 1 region that contribute to the early detection of the obstruction in the minor groove. Taken together, these studies highlight the importance of pol II recognition of DNA damage and modifications in the maintenance of transcriptional fidelity.

Recent earthquake disasters have demonstrated the seismic vulnerability of highway bridge systems and the significance of the ensuing social impact. Rapid seismic assessment of regional highway bridges is critical to help reduce the severe loss of life and property. Firstly, the typical modeling technique for reinforcement concrete highway bridges is introduced using specific elements for different components. However, the modeling procedures are material-level parameter-dependent and time-consuming. The nonlinear analysis convergence is also a frustrating problem for numerical simulation. Due to these realistic limitations, a simple, fast, and robust numerical model that can be developed with only component-level information needs to be adopted. It’s shown that the bridge bent representation can be simplified as a single degree of freedom system. The force-displacement relationship of the bridge can be roughly approximated by a bilinear curve. So a simplified 2D bilinear model is adopted for highway bridges throughout the study. Secondly, the statistical distributions for selected bridge input parameters can be derived based on the regional bridge inventory. Then an iterative process by sampling and filtering input parameters can be used to generate as many bridge candidates as possible for a specific region. The proposed bridge models and selected historical ground motions will be utilized to develop a seismic response prediction model using machine learning for an instrumented highway bridge. This study investigates the optimal features to represent the highway bridge and ground motion. Different regression models are applied for bridges with near-fault motion, and it’s shown the accuracy of the guided machine learning prediction model has exceeded the performance of traditional methods. A discounted accuracy is observed when applying the same machine learning prediction model in the highway bridge with far-field ground motion, And a two-layer LSTM network is developed with a new representation of far-field ground motion as the input.

Recent earthquake disasters have demonstrated the seismic vulnerability of highway bridge systems. Rapid seismic assessment of regional highway bridges is critical to help reduce severe loss of life and property. However, measurement of the regional scale system performance faces the challenge of dealing with the large uncertainty in structural properties and spatial characteristics. Traditionally, the numerical modeling approaches are established to simulate nonlinear response for each highway bridge across a regional portfolio. This process is largely limited by accuracy of model and computational effort. Especially some key structural component parameters are almost impossible to be retrieved for some ancient bridges. An alternative data-driven framework is developed to predict seismic responses or damage level of bridges using machine learning techniques. The proposed hierarchically structured framework enables a customized application in different scenarios. Firstly, the typical modeling technique for reinforcement concrete highway bridges is introduced using specific elements for different components. However, the modeling procedures are material-level parameter dependent and time consuming. The nonlinear analysis convergence is also a frustrating problem for numerical simulations. Due to these realistic limitations, a simple, fast and robust numerical model which can be developed with only component-level information needs to be adopted. It’s shown that the bridge bent representation can be simplified as a single degree of freedom system. The force-displacement relationship of the bridge can be roughly approximated by a bilinear curve. So a simplified 2D bilinear model is adopted for highway bridges throughout the study. Secondly, the statistical distributions for selected bridge input parameters can be derived based on the regional bridge inventory. Then an iterative process by sampling and filtering input parameters can be used to generate as many bridges as possible candidates for a specific region. The proposed bridge models and selected historical ground motions will be utilized to develop a seismic response prediction model using machine learning for instrumented highway bridges. This study investigates the optimal features to represent the highway bridge and ground motion. Different regression models are applied for near-fault motions and far-field motions and similar performance can be achieved, which significantly outperformed the traditional methods. Finally, to predict the seismic response of the non-instrumented highway bridges whose ground motion information is missing, the kriging interpolation model is implemented first. Then graph network is exploited to improve the performance. Different rules are evaluated for constructing an undirected graph for the highway bridges in an active seismic region. Subsequently, the Node2vec model is conducted to extract the embedding for each node and a graph neural network is implemented to predict the seismic response. Furthermore, vast amounts of text description data from online social platforms can be used to help detect the potential severely damaged bridges rapidly once an earthquake happens. A Convolution Neural Network classification model is implemented to evaluate the overall damage level distribution based on the collected text data. GloVe model is used to generate the word vector as its distributed representation.

Transcription Coupled Nucleotide Excision repair (TC-NER) is a widely conserved mechanism that is used to repair DNA damage and maintain genome stability by targeting bulky DNA lesions during transcription. In S. Cerevisiae, TC-NER is initiated by the recognition and binding of the Rad26 protein (or CSB in H.Sapiens) directly to bulky lesion stalled Pol II and upstream of the dsDNA fork.12 The Rad26 protein can discriminate between a bulky lesion stalled Pol II from other types of stalled Pol II, determining whether downstream TC-NER factors such as UVSSA and TFIIH are needed for recruitment.1 The absence or mutation of the Rad26 homolog, the CSB protein, causes severe genetic disease such as Cockayne Syndrome (CS), highlighting its functional importance.2 The function of the Rad26 is deemed critically important to understanding bulky DNA damage repair by TC-NER. While the structure of Rad26-Pol II complexes have been determined using Cryo-EM structure studies, the Rad26 NTD and CTD flanking regions are not revealed due to flexibility. These flanking regions connect to the core domain by a flexible loop and are important for Rad26 function via interaction with different protein partners. To investigate the unknown function of the Rad26 NTD and CTD, a GST pull-down assay is performed using GST-Rad26 fusion proteins in a S. Cerevisiae system to help determine potential protein binding factors that may bring insight on the function of these flanking regions of the Rad26. The pulled-down protein bands were analyzed using LC-MS. Potential binding partners that were discovered for the Rad26 NTD and CTD regions include Elf1, TFIIH subunit SSL1, FACT complex subunit POB3, and histone associate proteins RPD3, RVB2, and BDF1. The identification of these likely protein binding factors opens the possibility for future investigation on the protein partners of the Rad26 NTD and CTD. Insight on the proteomics of the Rad26 NTD and CTD interactome will help in understanding the roles of the Rad26 flanking regions in the TC-NER pathway.

During transcriptional elongation, RNA Polymerase (Pol II) may become stalled at DNA lesions. One way to resolve this transcriptional arrest is through Transcription Coupled-Nucleotide Excision Repair (TC-NER). Cockayne Syndrome Group B protein (CSB) is the first protein to be recruited and binds the upstream of stalled Pol II, alters the surrounding chromatin environment and allows other repair factors to access the DNA lesion. Many mutations in CSB are associated with Cockayne Syndrome (CS). Chapter 1 investigates the biochemical properties of CSB mutations observed in CS patients using Rhp26, the Schizosaccharomyces pombe ortholog of CSB. All six mutations studied show a decrease in chromatin remodelling and DNA translocation activities, both of which are important activities for CSB to execute its molecular function. In Chapter 2, we studied the regulation of CSB/Pol II interactions by its flanking regions using Rad26, the Saccharomyces cerevisiae ortholog of CSB. A conserved C-terminal region coupling motif promotes Rad26/Pol II interactions. Taken together, these results demonstrate that mutations to CSB that can be detrimental to its enzymatic activity and its role in maintaining genomic fidelity.

Approximately half of the world population suffers from iron and/or zinc deficiency, and millions suffer from protein-energy malnutrition, primarily from reliance on plant based staple foods. These foods are low in iron, zinc, and protein density relative to animal based foods. We and others are interested in genetic improvement of plants to increase the nutritional value of plants, a strategy termed biofortification. In previous work, the NAM transcription factor genes of wheat were shown to regulate leaf senescence and iron, zinc, and nitrogen remobilization and translocation from vegetative tissues to grain. Thus, genes of the NAM transcription factor regulon are potential targets for nutritional improvement of cereal or other seed crops. As a first step to identify NAM regulated genes, we used the Affymetrix Wheat Genome microarray to profile genes that are differentially regulated in flag leaf tissue at mid-grain fill relative to anthesis, and that are also differentially regulated between control and NAM RNAi knockdown lines. Over three hundred genes met the criteria to be potential NAM targets, several of which are annotated as coding for proteins that could be involved in nutrient transport or protein metabolism. A highly homologous NAM gene with developmentally regulated leaf expression similar to wheat NAM genes was cloned from Sorghum bicolor. Results of genome-wide bioinformatic and molecular screens to identify potential NAM regulated genes and putative NAM response elements in gene promoters will be presented.