Biochemistry, Biophysics, and Structural Biology Commons^™

Rnai-Mediated Gene Silencing In The Exotic Redbay Ambrosia Beetle, Xyleborus Glabratus, And Insect-Fungal Interactions Within The Laurel Wilt Complex, Morgan Christine Knutsen

Theses and Dissertations--Entomology

Laurel wilt disease (LWD) is a lethal vascular disease impacting lauraceous hosts caused by Harringtonia lauricola, the fungal symbiont of the redbay ambrosia beetle (Xyleborus glabratus Eichoff) (RAB) (Coleoptera: Curculionidae). LWD has caused tree mortality throughout the southeastern United States and is continuing to spread into new regions. Current management methods have not been successful in preventing spread, warranting investigation into innovative techniques including RNA interference (RNAi).

Elongation factor-1 alpha (ef1a) and actin (act) were established as stably expressed reference genes after exposing beetles to different photoperiod, temperature, and dsRNA exposure. After RAB oral …

The Development And Characterization Of Nanobodies Specific To Protein Tyrosine Phosphatase 4a3 (Ptp4a3/Prl-3) To Dissect And Target Its Role In Cancer., Caroline Smith

Theses and Dissertations--Molecular and Cellular Biochemistry

Protein Tyrosine Phosphatase 4A3 (PTP4A3 or PRL-3) is an oncogenic dual-specificity phosphatase that drives tumor metastasis, promotes cancer cell survival, and is correlated with poor patient prognosis in a variety of solid tumors and leukemias. The mechanisms that drive PRL-3’s oncogenic functions are not well understood, in part due to a lack of research tools available to study this protein. The development of such tools has proven difficult, as the PRL family is ~80% homologous and the PRL catalytic binding pocket is shallow and hydrophobic. Currently available small molecules do not exhibit binding specificity for PRL-3 over PRL family members, …

Investigating The Relationship Between Metabolic Reprogramming And Peripheral Cd4+ T-Cell Inflammation In Human Type 2 Diabetes Pathogenesis, Gabriella Kalantar

Theses and Dissertations--Microbiology, Immunology, and Molecular Genetics

Chronic, low-grade systemic inflammation rises in obesity and promotes type 2 diabetes (T2D). Circulating immune cells are key indicators of obesity and T2D pathogenesis. T cells outnumber monocytes, in blood, suggesting that T cells might fuel peripheral inflammation in obesity/T2D. Our lab’s work supports this idea by identification of a Th17 cytokine profile in T2D from T-cell stimulated peripheral blood mononuclear cells. Work described herein further supported this work by demonstrating that T cells dominate peripheral inflammation over monocytes across the spectrum of obesity and glycemic control. Our lab has also recently shown that inflammation changes during prediabetes (preT2D), identified …

Characterization Of The Function And Regulation Of The Hmpv Phosphoprotein, Rachel Thompson

Theses and Dissertations--Molecular and Cellular Biochemistry

Human metapneumovirus (HMPV) is a non-segmented, negative strand RNA virus (NNSV) that frequently causes respiratory tract infections in infants, the elderly, and the immunocompromised. Despite the initial identification of HMPV in 2001, there are currently no FDA approved antivirals or vaccines available. Therefore, understanding the mechanism of HMPV replication is critical for the identification of novel therapeutic targets. A key feature in the replication cycle of HMPV and other NNSVs is the formation of membrane-less, liquid-like replication and transcription centers in the cytosol termed inclusion bodies (IBs). Recent work on NNSV IBs suggests they display characteristics of biomolecular condensates formed …

A Characterization Of Key Residues In Class I Viral Fusion Proteins Important For Fusogenic Activity, Hadley E. Neal

Theses and Dissertations--Molecular and Cellular Biochemistry

Viral fusion proteins are critical for viral entry and subsequent infection. Class I fusion proteins are characterized by synthesis as an inactive precursor requiring cleavage by a host cell protease to become fusion competent. Though vaccine and antiviral therapeutic developments often target the fusion protein, questions surrounding cleavage dynamics and protein stability remain. The work presented in this dissertation investigates specific regions of three class I viral fusion proteins in an effort to identify key residues involved in proteolytic processing and membrane fusion.

The trimeric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein (S) mediates receptor binding, facilitates fusion …

Developing A Biocatalytic Toolbox To Aid In Understanding Nucleoside Antibiotics, Jasmine Brianna Woods

Theses and Dissertations--Pharmacy

Antibiotic resistance happens when bacteria develop the ability to survive medications that normally terminate them. Instead, these super germs are able to survive in the body and produce a community of antibiotic resistance germs which can cause human fatalities. It is important to discover and develop new compounds and molecules that will improve this clinical obstacle. This research focused on analyzing the biosynthesis that incorporates distinctive chemical characteristic of various nucleoside antibiotics, ß-hydroxy amino acids and α-methyl-amino acids. ß-hydroxy amino acids and α-methyl-amino acids are considered an important class of industrially useful compounds, particularly for pharmaceutical development, and are found …

Plant Cell Wall Composition And In Vitro Fermentation Characteristics Of Cool-Season Forage Grasses From Two Growing Seasons In Central Kentucky, Sophia Danielle Newhuis

Theses and Dissertations--Animal and Food Sciences

Grass cell walls are rich in cellulose, hemicellulosic arabinoxylan (AX) polysaccharides, and lignin. AX structural differences such as degree and pattern of branching and the ester-linked phenolic acid content could affect plants’ digestibility when used as forage for livestock. However, there is little information about how these structural elements change over the growing season in the vegetative tissue of cool-season perennial grasses. Enhanced information about the cell wall composition and carbohydrate structure of forage material will provide a foundation for expanding our knowledge of how forage cell wall carbohydrate structures are utilized by ruminants. The objectives of this study were …

Long-Read Sequencing Of The Zebrafish Genome Reorganizes Genomic Architecture, Yelena Chernyavskaya, Xiaofei Zhang, Jinze Liu, Jessica S. Blackburn

Molecular and Cellular Biochemistry Faculty Publications

BACKGROUND: Nanopore sequencing technology has revolutionized the field of genome biology with its ability to generate extra-long reads that can resolve regions of the genome that were previously inaccessible to short-read sequencing platforms. Over 50% of the zebrafish genome consists of difficult to map, highly repetitive, low complexity elements that pose inherent problems for short-read sequencers and assemblers.

RESULTS: We used long-read nanopore sequencing to generate a de novo assembly of the zebrafish genome and compared our assembly to the current reference genome, GRCz11. The new assembly identified 1697 novel insertions and deletions over one kilobase in length and placed …

Prostacyclin Promotes Degenerative Pathology In A Model Of Alzheimer’S Disease, Tasha R. Womack, Craig T. Vollert, Odochi Ohia-Nwoko, Monika Schmitt, Saghi Montazari, Tina L. Beckett, David Mayerich, M. Paul Murphy, Jason L. Eriksen

Molecular and Cellular Biochemistry Faculty Publications

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that is the most common form of dementia in aged populations. A substantial amount of data demonstrates that chronic neuroinflammation can accelerate neurodegenerative pathologies. In AD, chronic neuroinflammation results in the upregulation of cyclooxygenase and increased production of prostaglandin H2, a precursor for many vasoactive prostanoids. While it is well-established that many prostaglandins can modulate the progression of neurodegenerative disorders, the role of prostacyclin (PGI2) in the brain is poorly understood. We have conducted studies to assess the effect of elevated prostacyclin biosynthesis in a mouse model of AD. Upregulated prostacyclin expression …

Ppld Is A De-N-Acetylase Of The Cell Wall Linkage Unit Of Streptococcal Rhamnopolysaccharides, Jeffrey S. Rush, Prakash Parajuli, Alessandro Ruda, Jian Li, Amol Arunrao Pohane, Svetlana Zamakhaeva, Mohammad M. Rahman, Jennifer C. Chang, Artemis Gogos, Cameron W. Kenner, Gérard Lambeau, Michael J. Federle, Konstantin V. Korotkov, Göran Widmalm, Natalia Korotkova

Molecular and Cellular Biochemistry Faculty Publications

The cell wall of the human bacterial pathogen Group A Streptococcus (GAS) consists of peptidoglycan decorated with the Lancefield group A carbohydrate (GAC). GAC is a promising target for the development of GAS vaccines. In this study, employing chemical, compositional, and NMR methods, we show that GAC is attached to peptidoglycan via glucosamine 1-phosphate. This structural feature makes the GAC-peptidoglycan linkage highly sensitive to cleavage by nitrous acid and resistant to mild acid conditions. Using this characteristic of the GAS cell wall, we identify PplD as a protein required for deacetylation of linkage N-acetylglucosamine (GlcNAc). X-ray structural analysis indicates …

The Impact Of Plant Secondary Metabolites On Auxin And Cytokinin Signaling, Timothy E. Shull

Theses and Dissertations--Plant and Soil Sciences

Secondary metabolites are a broad class of specialized compounds that mediate plant-environment interactions and mitigate stress. It is increasingly clear that many phenylalanine-derived secondary metabolites are nearly indispensable for plant survival and that plants adjust their growth according to their secondary metabolic outputs. Consequently, many phenylalanine-derived secondary metabolites have influence over hormone activity. For instance, multiple phenylpropanoid intermediates and catecholamines alter the sensitivity of plants to the central hormone auxin, which in concert with cytokinin directs most aspects of plant growth and development. This dissertation reviews previous research on the influence of phenylpropanoid intermediates and catecholamines on plants, with a …

Apoe Genotype And Cerebral Glucose Metabolism: A Multi-Omics Approach, Holden C. Williams

Theses and Dissertations--Physiology

Apolipoprotein E (APOE) is encoded by the APOE gene, present in humans as three main isoforms (E2, E3, and E4). E4 carriers face up to a 15-fold increased risk for developing late-onset Alzheimer’s disease (AD), while E2 carriers are protected. Understanding the risk conferred by E4 has been an extensive research focus for nearly three decades, but the exact mechanism has yet to be proven. Many studies have demonstrated attenuated roles of E4 in classical hallmarks of AD, notably amyloid processing and neurofibrillary formation, which normally present later in disease progression. How APOE influences hallmarks that present much earlier are …

A Multidisciplinary Characterization Of The Enzymology And Biology Of Reversible Glucan Phosphorylation In Toxoplasma Gondii , Robert Murphy

Theses and Dissertations--Molecular and Cellular Biochemistry

Toxoplasma gondii is an opportunistic, protozoan parasite of all warm-blooded animals, infecting roughly one-third of humans worldwide. Humans acquire infections by consuming T. gondii tissue cysts in undercooked meat or from oocysts shed in cat feces. Encysted parasites convert into rapidly growing tachyzoites that disseminate throughout the body, defining the acute phase of infection. Under host immune pressure, tachyzoites convert into bradyzoites that populate tissue cysts found in CNS or muscle tissue and persist for the lifetime of the host, defining the chronic phase of infection. Tissue cysts are responsible for transmission via carnivory, but also possess the ability to …

Peripheral And Central Glucose Flux In Type I Diabetes, Jelena Anna Juras

Theses and Dissertations--Neuroscience

Diabetes is a complex metabolic disorder, of which high blood glucose concentration is the primary hallmark. Type I diabetes mellitus (T1DM) is characterized by the lack of insulin production, due to a poorly understood autoinflammatory cascade. In the words of historian Barnett “Diabetes may no longer be a death sentence, but for more and more people in the 21st century, it will become a life sentence”, making it the focal point of many research groups. It is estimated that around 20 million individuals worldwide live with T1DM.

Effects of long-term chronically elevated blood glucose are not only seen in micro/macro-vascular …

Mechanism Of Antibiotic Permeability Through The Gram-Negative Bacterial Envelope, Olaniyi Alegun

Theses and Dissertations--Chemistry

The outer membrane of Gram-negative bacteria (GN) makes them distinct among superbugs that are associated with the development of antibiotic resistance. The outer membrane, and inner membrane, separated by the periplasm, form a double-membrane barrier to the entry of antibiotics into the cell. Several studies have been conducted to examine the role of outer membrane modifications such as porins, lipopolysaccharides, and efflux pumps on antibiotic resistance. However, there is a paucity of knowledge on how antibiotics behave in the periplasm, to gain access into their target region. My thesis focuses on understanding the mechanism of antibiotic permeability through the cellular …

Building Tools For Improved Modulation Of The Human Gabaa Receptor, A Central Nervous System Target For The Treatment Of Anxiety, Garrett Edward Zinck

Theses and Dissertations--Pharmacy

In the U.S., anxiety is recognized as an increasing range of mentally and physically debilitating psychiatric health disorders with significant economic repercussions. Over the last 20 years, several novel anti-anxiety therapies have entered the drug development pipeline, but none have made it to market.

The work in this dissertation focused on structurally modifying valerenic acid (VA), a structurally unique carboxylated sesquiterpene acid found in Valeriana officinalis. VA is putatively reported to have allosteric modulatory activity of the human GABA_A receptor, a ligand-gated ion channel responsible for attenuating neurotransmissions. Structural modeling of VA’s GABA_A receptor interaction suggests that …

Synthesis Of 6,6- And 7,7-Difluoro-1-Acetamidopyrrolizidines And Their Oxidation Catalyzed By The Nonheme Fe Oxygenase Lolo, Nabin Panth

Theses and Dissertations--Chemistry

One of the remarkable steps in loline alkaloid biosynthesis is the installation of an ether bridge between two unactivated C atoms in 1-exo-acetamidopyrrolizidine (AcAP). LolO, a 2-oxoglutarate-dependent nonheme Fe oxygenase, catalyzes both the hydroxylation of AcAP and the resulting alcohol's cycloetherification to give N-acetylnornoline (NANL). The mechanism of hydroxylation is well understood, but the mechanism of the oxacyclization is not. I synthesized difluorinated analogs of AcAP in an attempt to further understand the mechanism of the unusual cycloetherification step.

I prepared 6,6-F₂-AcAP in eight steps from N,O-protected 4-oxoproline. The key step was a Dieckmann …

Structural Basis Of Bacterial Flagellin For Naip5 Binding And Nlrc4 Inflammasome Activation And The Mechanism Of Flagellin Induced Release Of Cytokines In Vivo, Jian Cui

Theses and Dissertations--Chemistry

The bacterial flagellum is a whip-like structure that protrudes from the cell membrane and is one of the most complex and dynamic biological molecular machines that propels bacteria to swim toward beneficial environments and the sites of infection. It is composed of a basal body, a hook, and a long filament. The flagellar filament contains thousands of copies of the protein flagellin (FliC) monomer arranged helically and ending with a filament cap composed of oligomer protein FliD. The overall structure of the filament core is preserved across bacterial species, while the outer domains exhibit high variability, and in some cases …

Antibiotic Permeation In Gram-Negative Bacteria And Contribution Of Inflammasome Activation And Pyroptosis In Pathogenesis Of Salmonella Systemic Infection, Ankit Pandeya

Theses and Dissertations--Chemistry

Antibiotic resistance is one of the major global issues in the field of public health and medicine. Good antibiotic candidates need to be selectively toxic, inhibit cellular target, and effectively penetrate and accumulate in bacterial cells. The last factor is a formidable barrier in the development of antimicrobials effective in Gram-negative bacteria, due to the presence of two layers of cell envelope. The first half of my thesis focuses on understanding the permeation of small molecules through this formidable cell envelope, distribution inside the cell of Gram-negative bacteria, and design of novel methods to make small molecules effectively cross the …

The Role Of Charge On Dna Packaging And Integrity Within Reconstituted Peptide-Dna Assemblies, Ehigbai Oikeh

Theses and Dissertations--Chemistry

In nature, DNA exists primarily in a highly compacted form. The compaction of DNA in vivo is mediated by cationic proteins; histone in somatic nuclei and arginine-rich peptides called protamines in sperm chromatin. The packaging in the sperm nucleus is significantly higher than somatic nuclei resulting in a final volume roughly 1/20^th that of a somatic nucleus. This tight packaging results in a near crystalline packaging of the DNA helices. While the dense packaging of DNA in sperm nuclei is considered essential for both efficient genetic delivery as well as DNA protection against damage by mutagens and oxidative species, …

Development Of Fluorescence Based Approaches To Understand Astrocyte Biology In The Context Of Nicotine And Nicotinic Receptor Activity, Surya P. Aryal

Theses and Dissertations--Chemistry

Smoking and tobacco use (STU) is a major global health problem and worldwide more than six million people die due to tobacco related diseases each year. Although majority of smokers try to quit smoking several times in their life, traditional therapeutic approaches, which focus only on neuronal cells, have a very low success rate. Understanding the effect of nicotine on glial cells, synaptic communication and blood vasculature in the brain can provide further insights on the neurobiology of substance abuse and can potentially help to design better therapeutic approaches. Glial cells are non-excitable cells in the brain which do not …

Late-Life Exercise Mitigates Skeletal Muscle Epigenetic Aging, Kevin A. Murach, Andrea L. Dimet-Wiley, Yuan Wen, Camille R. Brightwell, Christine M. Latham, Cory M. Dungan, Christopher S. Fry, Stanley J. Watowich

Center for Muscle Biology Faculty Publications

There are functional benefits to exercise in muscle, even when performed late in life, but the contributions of epigenetic factors to late-life exercise adaptation are poorly defined. Using reduced representation bisulfite sequencing (RRBS), ribosomal DNA (rDNA) and mitochondrial-specific examination of methylation, targeted high-resolution methylation analysis, and DNAge™ epigenetic aging clock analysis with a translatable model of voluntary murine endurance/resistance exercise training (progressive weighted wheel running, PoWeR), we provide evidence that exercise may mitigate epigenetic aging in skeletal muscle. Late-life PoWeR from 22–24 months of age modestly but significantly attenuates an age-associated shift toward promoter hypermethylation. The epigenetic age of muscle …

Analysis Of Hendra Virus Fusion Protein N-Terminal Transmembrane Residues, Chelsea T. Barrett, Hadley E. Neal, Kearstin Edmonds, J. Lizbeth Reyes Zamora, Carole L. Moncman, Andreea Popa, Everett Clinton Smith, Stacy R. Webb, Rebecca Ellis Dutch

Molecular and Cellular Biochemistry Faculty Publications

Hendra virus (HeV) is a zoonotic enveloped member of the family Paramyoxviridae. To successfully infect a host cell, HeV utilizes two surface glycoproteins: the attachment (G) protein to bind, and the trimeric fusion (F) protein to merge the viral envelope with the membrane of the host cell. The transmembrane (TM) region of HeV F has been shown to have roles in F protein stability and the overall trimeric association of F. Previously, alanine scanning mutagenesis has been performed on the C-terminal end of the protein, revealing the importance of β-branched residues in this region. Additionally, residues S490 and Y498 have …

Untargeted Lipidomics Of Non-Small Cell Lung Carcinoma Demonstrates Differentially Abundant Lipid Classes In Cancer Vs. Non-Cancer Tissue, Joshua M. Mitchell, Robert M. Flight, Hunter N. B. Moseley

Molecular and Cellular Biochemistry Faculty Publications

Lung cancer remains the leading cause of cancer death worldwide and non-small cell lung carcinoma (NSCLC) represents 85% of newly diagnosed lung cancers. In this study, we utilized our untargeted assignment tool Small Molecule Isotope Resolved Formula Enumerator (SMIRFE) and ultra-high-resolution Fourier transform mass spectrometry to examine lipid profile differences between paired cancerous and non-cancerous lung tissue samples from 86 patients with suspected stage I or IIA primary NSCLC. Correlation and co-occurrence analysis revealed significant lipid profile differences between cancer and non-cancer samples. Further analysis of machine-learned lipid categories for the differentially abundant molecular formulas identified a high abundance sterol, …

An Empirical Pipeline For Personalized Diagnosis Of Lafora Disease Mutations, M. Kathryn Brewer, Maria Machio-Castello, Rosa Viana, Jeremiah L. Wayne, Andrea Kuchtová, Zoe R. Simmons, Sarah Sternbach, Sheng Li, Maria Adelaida García-Gimeno, Jose M. Serratosa, Pascual Sanz, Craig W. Vander Kooi, Matthew S. Gentry

Molecular and Cellular Biochemistry Faculty Publications

Lafora disease (LD) is a fatal childhood dementia characterized by progressive myoclonic epilepsy manifesting in the teenage years, rapid neurological decline, and death typically within ten years of onset. Mutations in either EPM2A, encoding the glycogen phosphatase laforin, or EPM2B, encoding the E3 ligase malin, cause LD. Whole exome sequencing has revealed many EPM2A variants associated with late-onset or slower disease progression. We established an empirical pipeline for characterizing the functional consequences of laforin missense mutations in vitro using complementary biochemical approaches. Analysis of 26 mutations revealed distinct functional classes associated with different outcomes that were supported by clinical …

Apoε4 Lowers Energy Expenditure In Females And Impairs Glucose Oxidation By Increasing Flux Through Aerobic Glycolysis, Brandon C. Farmer, Holden C. Williams, Nicholas A. Devanney, Margaret A. Piron, Grant K. Nation, David J. Carter, Adeline E. Walsh, Rebika Khanal, Lyndsay E. A. Young, Jude C. Kluemper, Gabriela Hernandez, Elizabeth J. Allenger, Rachel Mooney, Lesley R. Golden, Cathryn T. Smith, J. Anthony Brandon, Vedant A. Gupta, Philip A. Kern, Matthew S. Gentry, Josh M. Morganti, Ramon C. Sun, Lance A. Johnson

Physiology Faculty Publications

BACKGROUND: Cerebral glucose hypometabolism is consistently observed in individuals with Alzheimer's disease (AD), as well as in young cognitively normal carriers of the Ε4 allele of Apolipoprotein E (APOE), the strongest genetic predictor of late-onset AD. While this clinical feature has been described for over two decades, the mechanism underlying these changes in cerebral glucose metabolism remains a critical knowledge gap in the field.

METHODS: Here, we undertook a multi-omic approach by combining single-cell RNA sequencing (scRNAseq) and stable isotope resolved metabolomics (SIRM) to define a metabolic rewiring across astrocytes, brain tissue, mice, and human subjects expressing APOE4.

RESULTS: Single-cell …

Awegnn: Auto-Parametrized Weighted Element-Specific Graph Neural Networks For Molecules., Timothy Szocinski, Duc Duy Nguyen, Guo-Wei Wei

Mathematics Faculty Publications

While automated feature extraction has had tremendous success in many deep learning algorithms for image analysis and natural language processing, it does not work well for data involving complex internal structures, such as molecules. Data representations via advanced mathematics, including algebraic topology, differential geometry, and graph theory, have demonstrated superiority in a variety of biomolecular applications, however, their performance is often dependent on manual parametrization. This work introduces the auto-parametrized weighted element-specific graph neural network, dubbed AweGNN, to overcome the obstacle of this tedious parametrization process while also being a suitable technique for automated feature extraction on these internally complex …

Hierarchical Harmonization Of Atom-Resolved Metabolic Reactions Across Metabolic Databases, Huan Jin, Hunter N. B. Moseley

Molecular and Cellular Biochemistry Faculty Publications

Metabolic models have been proven to be useful tools in system biology and have been successfully applied to various research fields in a wide range of organisms. A relatively complete metabolic network is a prerequisite for deriving reliable metabolic models. The first step in constructing metabolic network is to harmonize compounds and reactions across different metabolic databases. However, effectively integrating data from various sources still remains a big challenge. Incomplete and inconsistent atomistic details in compound representations across databases is a very important limiting factor. Here, we optimized a subgraph isomorphism detection algorithm to validate generic compound pairs. Moreover, we …

Effect Of Clinical Isolate Or Cleavage Site Mutations In The Sars-Cov-2 Spike Protein On Protein Stability, Cleavage, And Cell-Cell Fusion, Chelsea T. Barrett, Hadley E. Neal, Kearstin Edmonds, Carole L. Moncman, Rachel Thompson, Jean M. Branttie, Kerri Beth Boggs, Cheng-Yu Wu, Daisy W. Leung, Rebecca E. Dutch

Molecular and Cellular Biochemistry Faculty Publications

The trimeric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike protein (S) is the sole viral protein responsible for both viral binding to a host cell and the membrane fusion event needed for cell entry. In addition to facilitating fusion needed for viral entry, S can also drive cell-cell fusion, a pathogenic effect observed in the lungs of SARS-CoV-2-infected patients. While several studies have investigated S requirements involved in viral particle entry, examination of S stability and factors involved in S cell-cell fusion remain limited. A furin cleavage site at the border between the S1 and S2 subunits (S1/S2) has …

Algebraic Graph-Assisted Bidirectional Transformers For Molecular Property Prediction, Dong Chen, Kaifu Gao, Duc Duy Nguyen, Xin Chen, Yi Jiang, Guo-Wei Wei, Feng Pan

Mathematics Faculty Publications

The ability of molecular property prediction is of great significance to drug discovery, human health, and environmental protection. Despite considerable efforts, quantitative prediction of various molecular properties remains a challenge. Although some machine learning models, such as bidirectional encoder from transformer, can incorporate massive unlabeled molecular data into molecular representations via a self-supervised learning strategy, it neglects three-dimensional (3D) stereochemical information. Algebraic graph, specifically, element-specific multiscale weighted colored algebraic graph, embeds complementary 3D molecular information into graph invariants. We propose an algebraic graph-assisted bidirectional transformer (AGBT) framework by fusing representations generated by algebraic graph and bidirectional transformer, as well as …