136879 research outputs found
Sort by
Developing the Texas A&M Smart and Connected Homes Testbed
No full textThe Texas A&M Smart and Connected Homes Testbed has been developed to support residential HVAC research. The flexible testbed enables the windows and walls to be replaced, the floorplan to be reconfigured, provides two separate duct networks, and incorporates on-site renewable energy. Heavy instrumentation is done at the testbed to capture information on local weather conditions, building envelope performance, occupant comfort, HVAC equipment performance, and the power consumption of all household end-uses. A smart thermostat is also incorporated to provide HVAC control capabilities. Occupants are simulated inside the home to mimic actual operation and internal loads. A Modelica model has been created for the building envelope and split system HVAC at the testbed. Using data from the experimental testbed, the model is tuned to ensure accurate implementation. Researchers can use these models to bridge the gap between simulation-based studies and their real-world application
Geometric Deep Learning for Molecular Discoveries
No full textWith the rapid advancement of artificial intelligence (AI), its applications in scientific research have grown significantly, giving rise to the research area of AI for science (AI4Science). In this dissertation, we focus on AI for molecular science, such as small molecules and proteins, aiming to build efficient and effective methods to accelerate molecular discovery. we particularly focus on two fundamental tasks, molecular representation learning and molecule generation. Specifically, we model molecules as graphs and design geometric deep learning methods for molecules.
We first consider representation learning of 3D molecular graphs, where each node has its 3D coordinates. With an accurate representation learning model, we can reduce the computation time required for predicting molecular properties. In this dissertation, we provide an analysis in the spherical Coordinate System (SCS) for the complete identification of 3D graph structures and propose our SphereNet. SphereNet can distinguish similar molecular structures, such as two enantiomers that are mirror images of each other and reduce complexity from O(nk��) to O(nk��),
enabling it to perform efficiently on large-scale molecules. Here n and k denote the number of nodes and the average degree in the 3D graph, respectively.
While SphereNet presents advancements in accuracy and efficiency, it still can not incorporate 3D information completely. Furthermore, its complexity remains higher than some existing methods. We then propose ComENet to address these issues and incorporate 3D information completely and efficiently. Our method guarantees full completeness of 3D information on 3D graphs by achieving global and local completeness with a complexity of O(nk).
SphereNet and ComENet are tailored for small molecules. Extending their application to proteins is challenging due to the large number of atoms in proteins and their inherent multi-level nature. Therefore, we further design our method ProNet specifically for proteins. ProNet completely captures three levels of protein structures, e.g., the amino acid, backbone, or all-atom levels and is more efficient than existing methods. ProNet can be applied on various downstream tasks, including protein fold and function prediction, protein-ligand binding affinity prediction, and protein-protein interaction prediction.
Lastly, we consider 3D molecule generation. The generation of novel molecules with desired properties is an important step in drug discovery. In this dissertation, we apply language models (LMs) for 3D molecule generation by introducing our canonical and SE(3)-invariant tokenizer, Geo2Seq. Experiments show that our new method can achieve promising results
EMF Shielding of Stepper Motors as a Means of Improving the Security of CNC Operations
No full textThis paper is a documentation on a study conducted exploring possible shield samples that can be used on standardized stepper motors. The cause for this study is the existence of a security vulnerability in computer numerical control (CNC) operations in which the stepper motors that are used in this operation emit a distinctive electromagnetic field (EMF). This distinctive EMF can be captured by an EMF reader and recorded by an outside party to reproduce the operation. For private CNC operators, this is a source for an unwanted leak of data and needs to be addressed. As a preventative measure, it was proposed to explore ways to passively shield the motors in operations. Therefore, twenty-seven different shields of varying degrees of freedom were used as candidates to test the effectiveness of these degrees of freedom. The degrees of freedom include the material, infill geometry, and overall thickness. Additionally, measurements were made at two different distances from the motor to observe the shielding effects in both close range and long range. Aside from the degrees of freedom, the experiment was thoroughly controlled for accuracy. Measurements from the experiments were then collected and compiled for visual and statistical analysis. During statistical analysis, it was discovered that the data collected failed to meet the requirements of a parametric analysis leading to the use of a nonparametric means instead. The statistical analysis concluded the degrees of freedom that were specifically explored were statistically not significant. The results in this study lead to other possible factors that may prove to be more effective in shielding EMF. These other possible factors may lead to further work to be conducted outside of the scope of this thesis
Computational Biomechanics for a Standing Human Body: Modal Analysis and Simulation
No full textWe develop computational mechanical modeling and methods for the analysis and simulation of the motions of a human body. This type of work is crucial in many aspects of human life, ranging from comfort in riding, the motion of aged persons, sports performance and injuries, and many ergonomic issues.
A prevailing approach for human motion studies is through lumped parameter models containing discrete masses for the parts of the human body with empirically determined spring, mass, damping coefficients. Such models have been effective to some extent; however, a much higher-fidelity modeling method is to model the human body as it is, namely, as a continuum.
We present this approach, and for comparison, we choose two digital CAD models of mannequins for a standing human body, one from the versatile software package LS-DYNA and another from open resources with some of our own adaptations. Our basic view in this paper is to regard human motion as a perturbation and vibration from an equilibrium position which is upright standing. A linear elastodynamic model is chosen for modal analysis, but a full nonlinear viscoelastoplastic extension is possible for full-body simulation.
The motion and vibration of these two mannequin models is analyzed by modal analysis, where the normal modes of motion are determined. LS-DYNA is used as the supercomputing and simulation platform. Four sets of low-frequency modes are tabulated, discussed, visualized, and compared. Higher frequency modes are also selectively displayed. We have found that these modes of motion and vibration form intrinsic basic modes of biomechanical motion of the human body. This view is supported by our finding of the upright walking motion as a low-frequency mode in modal analysis. Dynamic motions of CAD mannequins are also simulated by drop tests for comparisons and the validity of the models is discussed through Fourier frequency analysis. In the low-frequency range, our numerical results have provided a satisfactory self-consistent match as validation. All computed modes of motion are collected in several sets of video animations for ease of visualization. Samples of LSDYNA computer codes are also included for possible use by other researchers
Characterizing Membrane Protein-Lipid Interactions by Native Mass Spectrometry
No full textNative mass spectrometry (MS) is a powerful tool for quantitatively characterizing protein complexes and the interactions between protein and ligands due to the capability of preserving non-covalent interactions during measurement. However, preserving non-covalent interactions in native MS studies for membrane protein complexes can be challenging. Higher activation energy is usually needed for desolvation and detergent release, resulting in higher charge states on membrane protein, disrupting the tertiary structure and non-covalent interactions. Therefore, reducing the charges carried on membrane protein is crucial for native MS study. A series of distinct charge-reducing molecules, polyamines, were investigated for their application in reducing charges on membrane protein. The results indicate that polyamines exhibit enhanced charge-reduction potency, presenting innovative strategies to modulate charge states and preserve non-covalent interactions during native MS studies.
In addition to discovering charge-reducing molecules, native MS was applied to characterize bacterial ATP-binding cassette (ABC) transporter MsbA, a crucial player in bacteria lipopolysaccharides (LPS) biogenesis, and its interactions with lipids. This study discovered the binding of copper (II) to MsbA that modulates MsbA-lipid interactions, with atomic structure resolved by X-ray crystallography. In addition, the results of this study revealed the conformation-dependent lipid binding affinities of MsbA by native MS, especially for the LPS precursor, 3-deoxy-D-manno-oct-2-ulosonic acid (Kdo)2-lipid A (KDL). This finding from native MS guided the structural biology study that resolved a 3.6 ��-resolution structure of MsbA in an open, outward-facing conformation, revealing previously undiscovered KDL binding sites that are important for the functions of MsbA.
This study also explored the thermodynamics of the interactions between MsbA and KDL. Despite identifying two distinct LPS binding sites on MsbA, the thermodynamic basis for the interactions of MsbA-KDL remained unclear. Native MS revealed that KDL binding to MsbA is mainly driven by entropy. Basic residues contribute to the binding of KDL through positive coupling entropy, overcoming unfavorable coupling enthalpy. These findings indicated the effect of solvent reorganization, specifically the desolvation of lipid binding sites and the lipid headgroup, in driving KDL binding to MsbA. This study provides new insights into thermodynamic contributions from residues in membrane proteins to lipid binding
John Bickham field notebook: AK8501-AK9000.pdf
No full textBound book, each page corresponds to a karyotype slide data.Data pages for AK9001-AK9500 corresponding to unique identifiers of specimens/samples examined for biological research. Specimens are primarily housed at Texas A&M University; Biodiverstiy Research and Teaching Collection
Machine Learning-Based Automated Fault Detection and Diagnostics in Building Systems
No full textAutomated fault detection and diagnostics (AFDD) analysis in commercial building systems using machine learning (ML) can improve the building���s efficiency and conserve energy costs from inefficient equipment operation. Boolean rules-based analysis is standard in current AFDD solutions but limits analysis to the rules defined and calibrated by energy engineers. As part of this dissertation, an automated process was developed to provide ML-based building analytics to building engineers and operators with minimal training in ML. The process can be applied to buildings with a variety of configurations, which reduces time and manual effort required for fault analysis when compared to Boolean rule-based systems. The developed procedure introduces advanced diagnostics with automatically generated metrics to validate the ML model���s predictions and rank detected faults in order of fault severity. Explanations of the methodology used for the ML analysis include a description of the algorithms used.
The analysis was applied to a building on the Texas A&M University campus where the results are shown to illustrate the performance of the process using measured data from a commercial building. Three case studies which analyze the building���s equipment are presented to show ML���s advantages over rule-based analysis. ML can detect faults in the system caused by degrading components. ML can also detect faults in system components with missing sensors by modeling expected system operation and making comparisons to actual system operation. An example of ML detecting a failure in a building is shown along with a demonstration of the decision boundaries of ML-based FDD in comparison with Boolean rule-based analysis. The results from these examples are used to show the strengths and weaknesses of using ML for AFDD analysis
Optimizing Infectious Disease Control: Strategies for Social Separation and Vaccine Allocation with Equity Consideration
No full textAs global threats from infectious diseases intensify, as highlighted by the COVID-19 pandemic, the urgency to enhance control measures becomes evident. Strategies such as social separation and vaccine allocation are pivotal in disease management. However, despite extensive research, many models still fall short. They rely on oversimplified assumptions such as population homogeneity and a single-wave pattern of infection spread. Current research still lacks comprehensive approaches in addressing the effective and equitable implementation of these strategies in more complex, realistic scenarios.
To address these gaps, we explore various models for social separation and vaccine distribution, considering individual-specific factors and the multi-wave nature of pandemics observed in reality. Focusing on efficiency and equity in disease mitigation, we uncover key components of an optimal infectious disease control strategy. These insights help us to formulate effective algorithms, understanding the trade-offs between efficacy, costs, and fairness. Our work also leads to the development of an efficient, fair clustering algorithm that not only performs well in our disease mitigation context but also excels across various datasets.
Case studies underscore the advantages of strategies tailored to specific individual information and dynamic behavior. Such approaches are more effective, cost-efficient, and equitable than traditional disease control measures. The fair clustering algorithm we developed further demonstrates its advantages over benchmark algorithms
Characterizing Membrane Protein-Lipid Interactions by Native Mass Spectrometry
No full textNative mass spectrometry (MS) is a powerful tool for quantitatively characterizing protein complexes and the interactions between protein and ligands due to the capability of preserving non-covalent interactions during measurement. However, preserving non-covalent interactions in native MS studies for membrane protein complexes can be challenging. Higher activation energy is usually needed for desolvation and detergent release, resulting in higher charge states on membrane protein, disrupting the tertiary structure and non-covalent interactions. Therefore, reducing the charges carried on membrane protein is crucial for native MS study. A series of distinct charge-reducing molecules, polyamines, were investigated for their application in reducing charges on membrane protein. The results indicate that polyamines exhibit enhanced charge-reduction potency, presenting innovative strategies to modulate charge states and preserve non-covalent interactions during native MS studies.
In addition to discovering charge-reducing molecules, native MS was applied to characterize bacterial ATP-binding cassette (ABC) transporter MsbA, a crucial player in bacteria lipopolysaccharides (LPS) biogenesis, and its interactions with lipids. This study discovered the binding of copper (II) to MsbA that modulates MsbA-lipid interactions, with atomic structure resolved by X-ray crystallography. In addition, the results of this study revealed the conformation-dependent lipid binding affinities of MsbA by native MS, especially for the LPS precursor, 3-deoxy-D-manno-oct-2-ulosonic acid (Kdo)2-lipid A (KDL). This finding from native MS guided the structural biology study that resolved a 3.6 ��-resolution structure of MsbA in an open, outward-facing conformation, revealing previously undiscovered KDL binding sites that are important for the functions of MsbA.
This study also explored the thermodynamics of the interactions between MsbA and KDL. Despite identifying two distinct LPS binding sites on MsbA, the thermodynamic basis for the interactions of MsbA-KDL remained unclear. Native MS revealed that KDL binding to MsbA is mainly driven by entropy. Basic residues contribute to the binding of KDL through positive coupling entropy, overcoming unfavorable coupling enthalpy. These findings indicated the effect of solvent reorganization, specifically the desolvation of lipid binding sites and the lipid headgroup, in driving KDL binding to MsbA. This study provides new insights into thermodynamic contributions from residues in membrane proteins to lipid binding
Functional Analyses of TMTC-Type O-Mannosyltransferases in Drosophila
No full textOxygen-linked mannosylation, also known as Protein O-mannosylation, is a type of glycosylation, obstruction of which has been shown to cause severe phenotypes in humans, from neurological abnormalities to congenital muscular dystrophies. O-mannose modifications are of special interest due to their essential role in nervous system development. Protein O-mannosyltransferases (POM) are the enzymes responsible for O-mannosylation. They are highly conserved in metazoans. A majority of human O-mannosyltransferase enzymes have homologs present in Drosophila, including canonical protein O-mannosyltransferases, POMT1 and 2, (designated as Rt and Tw in Drosophila), and recently discovered non-canonical O-mannosyltransferases, transmembrane O-mannosyltransferases targeting cadherins 1-4 (TMTC1- TMTC4), all of which share conserved sequence, and structure with their human counterparts. TMTCs have been shown to selectively modify cadherins and protocadherins in humans. Yet the functions of the O-mannose modifications from individual TMTCs in humans and Drosophila are not well understood. In order to address this, we focus on elucidating the molecular mechanisms and functions of O-mannosylation mediated by TMTC1 & 2 in Drosophila. Using the advantages of the Drosophila model, I investigate the effect of TMTC-mediated modifications on the development of the nervous system, behavior, and neurological functions. In my project, I focus on a deeper understanding of how O-mannose modifications affect cadherin function at the molecular, cellular, and organismal levels. Investigating the localization, expression levels, and functional effects of TMTC targets in TMTC mutants will shed light on the function of O-mannose. I analyzed mutant alleles of TMTC1 and TMTC2 and employed rescue constructs to restore functions using both Drosophila and human constructs. I investigated possible redundancy and collaboration within the TMTC gene family by combining the downregulation of different TMTCs and using various transgenic approaches. My results revealed that TMTC1 & TMTC2 function in a partially redundant manner to establish sensory axon connections in the larval brain. My results shed light on the in vivo functions of TMTCs, their role in the regulation of cadherin functions, and have built a Drosophila model for further elucidation of the mechanism of thus highly conserved non-canonical O-mannosylation pathway in animals, including the role of TMTC genes in human biology and pathological conditions