DSpace@RPI (Rensselaer Polytechnic Institute)
Not a member yet
6809 research outputs found
Sort by
Electronic and vibrational properties of carbon nanostructures
No full textDecember 2022School of ScienceIn this thesis, first-principles calculations based on density functional theory (DFT) are used to study the electronic and vibrational properties of 2D, 1D, and 0D carbon nanostructures. Newly proposed carbon nanomaterials with tunable properties are potential candidates of sil- icon replacement in nanoelectronic applications. Part of studying new theoretical materials is assessing their dynamical stability, which is verified through performing phonon calculations on different investigated structures. Patterning 2D carbon allotropes into quasi-1D nanorib- bons is a broadly adopted strategy to open an electronic band gap, which is a desirable feature in areas such as field-effect transistors and switching devices. DFT calculations are performed on 1D tripentaphene nanoribbons to show that some of the investigated ribbons have semiconducting properties unlike their corresponding 2D tripentaphenes. Additionally, a DFT study on 2D naphthylene-β structure shows that the system exhibits semiconducting properties in its spin-polarized state due to its bipartition nature, where in a previous study, the system was confirmed to be metallic in its spin-paired state.Raman scattering is a reliable and non-destructive technique for the characterization of materials. As the graphene nanoribbons are transferred from their growth substrate to the device substrate, characterization of the ribbons is needed to identify their lengths and any defects as part of testing the functionality of the device. Raman spectra of 17-armchair graphene nanoribbons (17-AGNRs) are computed to study the effect of the length and width of the ribbons on the frequencies of the vibrational modes. The computed Raman spectra are then compared with experimental data, where the length and edge type of the ribbons measured in the experiment are identified.
Finally, a computational method for the calculation of tip-enhanced Raman scatter- ing (TERS) is developed, where the target molecule is placed in the proximity of a noble- metal nanoparticle at a probing tip leading to the enhancement of the Raman signal. This method uses the discrete-dipole approximation (DDA) to compute the optical response of the nanoparticle, while the polarizabilty of the target molecule is calculated by using the bond polarizability model. TERS spectra are calculated for biphenyl and buckminster- fullerene molecules, and then compared with experimental data confirming the reliability of the method. It is shown that as the tip get sharper, and the incident field is in resonance with the nanoparticle, the TERS effect is more pronounced. Finally and more interestingly, 2D and 3D plots showing scans of the molecule are obtained, which provide visualization of the structure in real space as well as its constituents and chemical properties on the molecular level.Ph
Autonomous controls for active space debris removal
No full textMay 2023School of EngineeringSpace debris poses a major threat to the long-term endeavors of humanity. A build-up and consequential cascade of debris collisions may very well render space travel nearly to completely impossible. The economic, scientific, and militaristic ramifications to the congestion of orbital planes around the Earth are very real. At the date of writing, several space agencies have acknowledged that active debris removal (ADR) is the only viable method to keep our near Earth orbits clear. Proposed here are two algorithms that will be on-board a 3U CubeSat, OSCaR (Obsolete Spacecraft Capture and Removal), that will track and de-orbit pieces of debris on orbit. The first algorithm is an improvement on standard linear quadratic regulator (LQR) attitude controls for CubeSats. Instead of a single linearization of the non-linear dynamics with corresponding LQR gains, OSCaR will relinearize and calculate the newly corresponding gains. This allows for a faster pointing time when near low angular velocities. In addition, OSCaR will also account for the dynamics of the reaction wheels--a consideration rarely taken into account on-board CubeSats.
The second algorithm proposed is an inverse kinematics based impulse method to bring OSCaR to a near-rendezvous. OSCaR must pass by the debris with a defined separation. It need not stop, but rather pass close enough to launch a net with an electromagnetic tether attached. Current literature either utilizes inefficient continuous controls to rendezvous or ignore the error accumulated by the linearization of the non-linear equations of motion. However, this impulsive method will subtract out predicted error at the start of the intercept maneuver to produce a highly accurate interception.Ph
A variational bayes approach to inferring neuronal network connectivity
No full textDecember 2022School of ScienceUncovering a neuronal circuit's structure is an important step toward developing a mechanistic understanding of that circuit's function. In this thesis, we develop an algorithm which can accurately reconstruct an integrate-and-fire neuronal network's connectivity from observations of its spike-train data. Our network reconstruction algorithm is designed to achieve two primary goals: efficiently recover network connectivity given minimal observed data, and quantify our uncertainty about each connection in the network. Pursuant to our first goal, we define a novel Bayesian statistical model for neuronal spike train data, uniquely structured around empirically observed characteristics of spike-trains from conductance based integrate-and-fire neuronal networks. In particular, the model directly accounts for the spike-reset-refractory-period dynamics characteristic of spiking neurons; additionally, the model's dimensionality is reduced by flexibly encoding the typical response a neuron has to presynaptic spikes. We show that building a model which accounts for the structure inherent in neuronal network spiking dynamics makes recovering network connectivity more data-efficient, especially for large networks. Furthermore, our Bayesian model has parameters which directly correspond to structural properties of the observed neuronal circuit. Most importantly, the model contains a connection-type parameter for each pair of observed neurons , which indicates if neuron is excitatorily presynaptic, inhibitorily presynaptic to, or disconnected from neuron . Thus, the posterior distribution over these parameters facilitates direct inference of the network's connectivity, direct inference of the connection type (excitatory or inhibitory) between each connected pair, and local, pairwise estimates of our uncertainty about each inferred connection. An approximation to the Bayesian model's posterior distribution is derived using a variational Bayes algorithm which is carefully designed to retain meaningful probabilities over the model's connection type parameters.Ph
Studies of amyloid fibril formation and the characterization of amyloid - gfp binding
No full textMay 2018School of ScienceProteins are complex heteropolymers that fold to a native structure encoded by its amino acid sequence. This process is often in competition with aggregation – the intermolecular association of protein to form insoluble aggregates. One type of aggregate are amyloid fibrils - long thin, highly ordered β-sheet aggregates that can form from a variety of protein sequences. The presence of amyloid fibrils in human tissue are linked to numerous diseases (i.e. Alzheimer’s, Parkinson’s, systemic amyloidosis etc.) and often their role in the disease pathology is not clear. the amyloidogenic protein PAPf39 is present as amyloid fibrils in the semen of healthy men. These fibrils have been shown to greatly increase the infectivity of HIV in vitro and are a subject of ongoing study. Previous work has shown that the formation of PAPf39 is pH dependent, requiring a pH above 5.5 for fibril formation to occur. Removal of the N-terminus of PAPf39 allows fibril formation to occur under both acidic and neutral conditions. Evidence shows that the pH dependence of fibril formation can be explained by a perturbation of the PAPf39 structural ensemble. NMR data in the form of coupling constants and chemical shifts used in conjunction with REMD simulations of monomeric PAPf39 reveal that changes in pH induce the formation of a new β-hairpin in the N-terminus. In addition, neutral pH conditions promote the formation of an α helix near residue 23. Discovering novel binding partners for amyloid fibrils is a subject of great interest. Such compounds could be used to disaggregate, remodel or inhibit the formation of disease-relevant amyloid fibrils. Previous work has shown that sfGFP (superfolder Green Fluorescent Protein) is capable of binding to many different types of amyloid fibril. GFP is an 11 strand β-barrel protein that is fluoresces green upon exposure to blue light. How sfGFP is able to bind to amyloid fibrils is largely unknown. Paramagnetic relaxation enhancement experiments reveal the presence of a binding site centered around three solvent exposed tyrosine residues. Removal of the tyrosine residues, as well as the location of the tyrosine residues on the β-barrel impacts the affinity and the binding mode of GFP for PAPf39 fibrils. Additionally, a novel equilibrium mode assay has been developed for measuring GFP-fibril binding that has potential for examining the affinity of GFP for many other fibril systems.Ph
Fabricating heterogeneous alginate microbead arrays with spatially prescribed distributions using laser direct-write bioprinting
No full textAugust 2023School of EngineeringManufacturing 3D tumor systems with control over the spatial placement of disparate cell types (e.g., tumor, stromal) has posed a significant challenge. Compared to other fabrication methods, bioprinting approaches are particularly well suited to the task because they offer the fabrication of high-resolution constructs with high spatial control. Inkjet bioprinting represents the benchmark bioprinting approach because it can print multiple live cells and rapidly generate intricate patterns with high-throughput fabrication. However, inkjet bioprinting has limited resolution, increased potential for clogging cell-loaded bioinks, and could potentially introduce high shear stresses to cells. Alternatively, LDW bioprinting enables nozzle-free, noncontact fabrication of cells and cell-loaded microbeads with high resolution and spatial control. Further, LDW can create and pattern size-controlled 3D cell microenvironments (i.e., microbeads, microcapsules) in a single step. Herein, this work addresses the fabrication of spatially heterogeneous microbead arrays across multiple bioinks via LDW bioprinting. Two methods for bioprinting multiple bioinks onto the same substrate were explored, utilizing untagged and fluorescent-tagged alginates, demonstrating potential approaches to fabricate spatially heterogeneous constructs. LDW showed accurate and precise microbead placement for creating constructs with prescribed spatial composition. Further, spatially heterogeneous constructs were fabricated with individual microbead placement and guided by either user-defined idealized geometries or histologic image templates. This platform provides critical foundations for fabricating heterogeneous 3D tumor models that better reflect patient heterogeneities.M
A computational approach to lexical semantic shift across time and domain : methods and applications
No full textDecember 2022School of ScienceNeural natural language models are designed to learn word and sequence representations from large volumes of text. Such amount of data is typically achieved by merging multiple heterogeneous corpora from the Web.
However, language use is entrenched in the social context it appears, and linguistic variations manifest social differentiation such as ethnicity, gender, sex, and social class.
Words may have their meanings altered based not only on the lexical context but also in the social context they emerge, being associated with the group or community who utilizes them.
These changes are the object of study of computational semantic shift methods, the majority of which are currently designed to handle temporal language change, or linguistic evolution, with little endeavor made towards characterizing changes across domains. In this work, we proposed a method to improve the current semantic shift techniques in cross-domain tasks, and demonstrated its capability in unsupervised feature learning tasks. We focused on addressing the two major challenges of this problem: the assumption of gradual language change used in temporal analysis, and the lack of labeled data for supervised learning.
In particular, we designed a self-supervised learning method to obtain monolingual mappings of words, and showed that it surpasses the performance of state-of-the-art baselines both on over time and cross-domain detection.
Moreover, we designed a framework for the explainability of semantic shifts based on the learned mappings, showing the words that are semantically shifted across input sources, explaining the shift via word representatives and examples in sentence. Finally, we confirmed that semantic shift is able to perform domain differentiation by applying it in a study of scientific news source credibility. The study showed that by using semantic shift in conjunction with citation and copy behavior as measures of concordance of news sources, we could learn representations that capture relevant information about them, such as credibility and political bias, creating clusters of sources that share similar traits.
A qualitative analysis of the observed clusters using semantic shift allowed us to characterize clusters of political conspiracy theorists and sources that propagate pseudoscience/health conspiracy theories.Ph
Insights into the purification of bispecific antibodies on multimodal systems: from chromatography to biophysics
No full textAugust 2021School of EngineeringBispecific antibodies (bsAb) have gained significant importance recently due to their ability to bind to multiple epitopes resulting in an increased efficacy towards their targeted indication as compared to monoclonal antibodies (mAb). While several techniques have been developed to steer the formation of the bsAb molecule, the efficient production of bsAbs is impacted by the inevitable formation of product related variants due to the mispairing of the light and heavy chains in the expression system. Due to similarities in structure and surface properties between the correctly formed bsAb and the mispaired variants, the effective removal of these impurities is extremely challenging with single mode resin systems. Multimodal resin systems can offer increased selectivity over single mode systems for similar molecules due to the unique combination of charge and hydrophobic moieties on the same ligand. The work presented in this thesis aims to shed light on the use of multimodal resins for the purification of bsAb from their impurities. To that end, a systematic workflow was developed and implemented to understand selectivity differences and preferred binding patches for bsAbs and their parental mAbs on a homologous set of multimodal cation exchange (MM CEX) resin systems. This workflow incorporated chromatographic screening of the parent mAbs and their fragments at various pH conditions followed by surface property mapping and protein footprinting using covalent labeling followed by LC/MS analysis.For the first set of molecules, linear gradient experiments on MM CEX resins showed that the bsAb molecule exhibited a unique transitory behavior on some of the resins as a function of pH. This lead to the hypothesis that specific domains of the molecule could be involved in preferential binding to the resin surface at different pH conditions. Domain contribution experiments indicated that the retention of both parental mAbs was likely driven by (Fab)2 interactions at higher pH conditions. Protein footprinting experiments performed at pH 7.5 using sulfo-nhs-acetate to identify the preferred binding patches for the parental mAbs and the bsAb confirmed that the Fc was indeed not involved at the binding interface and that residues in the variable regions of the Fabs were driving the interactions with the MM CEX resins. Parent mAb A containing more positively charged lysine residues in the variable region of the Fab as compared to Parent mAb B resulted in Parent mAb A being more retained on the MM CEX resins. While some of the same residues that were seen to be important for the parental mAbs were also found to be important for the bsAb, it is likely that different avidity contributions from the two Parental Fab arms of the bsAb played a key role in the observed chromatographic behavior.
This workflow was then extended to study the chromatographic behavior of a product related variant found in the bsAb expression pool. This product related variant comprising of heterodimerized heavy chains originating from both the parent mAbs and light chains originating only from Parent mAb B was found to have similar chromatographic behavior as Parent mAb B on most MM CEX resins. Protein footprinting experiments at pH 7.5 using sulfo-nhs-acetate indicated that that the same residues in the variable regions of the light and heavy chains of the individual parental mAbs were also important to the binding of this product related variant. The relative contributions to binding from the two Fab arms was again found to have a significant impact on the observed chromatographic behavior. Results from this study demonstrated why the purification of the correctly formed bsAb from the mispaired product related variants is such a challenge in downstream processing.
The workflow was then modified to include the use of a novel labeling chemistry, diethyl pyrocarbonate (DEPC). Unlike sulfo-nhs-acetate which required basic pH conditions limiting our studies, DEPC enabled us to perform the covalent labeling experiments at acidic pH conditions allowing for the study of preferred binding patches at lower pH conditions. Linear gradient experiments for the intact Parent mAb B and its fragments had indicated that at pH 5.5, the Fc may also be contributing to the chromatographic behavior of the mAb. Protein footprinting studies at pH 5.5 indicated that in addition to the contributions from the (Fab)2 domain, the Fc domain indeed played an important role in the retention of mAbs. This added avidity contribution from the Fc resulted in an increased retention of the molecule. In order to validate the chromatographic experiment for the constituent fragments, labeling studies were also performed for the individual fragments of mAb B at pH 5.5. Results indicated that while the residues found to be important for the binding of the intact mAb were also important for the retention of the fragments, additional residues that were sterically hindered on the intact mAb may be contributing to the binding interface of the fragments.
Finally, this workflow was extended to study the chromatographic behavior for an additional set of industrially important bsAb and it’s parent mAbs. The two bsAb share a common parental mAb resulting in a unique opportunity to study how the common half between the two bsAb impacts their chromatographic behavior. The results indicated that subtle differences in the charge and hydrophobicity landscape of the protein surface resulted in unique chromatographic selectivities for the parental molecules giving rise to opportunities for their separation using MM CEX resins. Further, initial chromatographic screening for the bsAb indicated opportunities for their separation. The work presented in this this thesis lays the foundation to systematically study the chromatographic selectivity of large multidomain molecules such as mAbs and bsAbs which can provide important insights into improved biomanufacturability and expedited downstream bioprocess development.Ph
Theorizing in a society of nature : a grounded theory analysis of permaculture online and textual discourse
No full textDecember 2022School of Humanities, Arts, and Social SciencesThis project explores conceptual themes in the ecological design movement known as permaculture. The project is guided by the question: How does the permaculture movement form its theoretical trajectories in service of achieving ecological repair? Within the movement, practitioners of permaculture interpret and apply insights provided by the founders of the program in what have become the core texts of the movement. These “design manuals” function as a type of instructional diagram, philosophical guide, and space of contestation for participants in the movement. The manuals are extensive, complex, and interpreted in unique and heterogenous ways by the broader community. To this point, scholarly articles have based their examinations of permaculture on these texts while ignoring the interpretive activities of contemporary practitioners. This dissertation attempts to resolve this gap in scholarship by exploring the online community of permaculture practitioners on the web forum Permies.com. Using grounded theory qualitative methods, the forum provides Permie's own voices to explain the conceptual underpinning of the movement. This research, then, exposes critical debates in the movement surrounding how participants negotiate meaning and attempt to build a permaculture worldview. This research identifies their use of system thinking, debates between spirituality and science discourse internal to the movement, and the social complexity surrounding the broader use of indigenous design techniques. This research also discovered that through these topics the community images a type of communication with Nature resulting in the movement's attention to what they call “pattern language.” For Permies pattern language is expressed as a type of embodied eco-literacy resulting from a complex relationship dynamic between practitioner and landscape. Additionally, this study highlights the intersection of permaculture theory with emerging theories of Other-than-human and nomadic design. Broader implications of the study contribute to debates of naturecultures and the construction of ecological based communities in online spaces. These findings provide deeper understanding of the ways in which ecological based subcultures self-create unique theoretical assemblies impacting participants' worldviews.Ph
First-order methods for large-scale distributed nonconvex optimization
No full textMay 2023School of ScienceDistributed optimization has garnered much attention from the machine learning community in recent years due to the ever-growing presence of distributed data and the emphasis on more complex machine learning models which require vast amounts of data to train. This work focuses on designing and analyzing first-order methods to solve distributed nonconvex optimization problems over a network of N computing devices (e.g. cell phones, GPUs). Specifically, we propose two general algorithmic frameworks: one for handling deterministic (offline) problems and another for handling stochastic (online) problems. In both cases, we rigorously prove our frameworks achieve optimal (full or sample) gradient complexity and in the deterministic setting we further achieve optimal communication complexity.Ph
A Concise Ontology to Support Research on Complex, Multimodal Clinical Reasoning
No full textWhen clinicians perform tasks involving clinical reasoning, such as the diagnosis or treatment of diabetes, multiple forms of reasoning, including deduction and abduction, are often employed. Ontologies designed to provide a foundation for clinical decision support systems have been encoded based on Clinical Practice Guidelines. Nevertheless, existing approaches solely allow deductive rules for clinical reasoning, with ontologies too large or complex to support tractable abductive reasoning. We follow existing guidelines and standards to design the Diabetes Pharmacology Ontology, a concise ontology – an ontology engineered by adhering to the Minimum Information to Reference an External Ontology Term principle and following an agile design approach. We claim that use cases that incorporate multiple forms of reasoning, such as those aimed at supporting both deduction and abduction, are better supported by concise, rather than complete and comprehensive, ontologies. We demonstrate how Personal Health Knowledge Graphs have been implemented using our ontology and evaluate the abductive capability of modules included with our ontology. We openly publish the resources that have resulted from this work, as listed below. This work demonstrates how multimodal semantic reasoning – deduction and abduction – can be used to emulate tasks involving clinical reasoning and thus has the potential to support practitioners with clinical decision-making