48440 research outputs found
Sort by
Phosphomimetic Mutation Alters the Kinetics of S. cerevisiae Malate Dehydrogenase 2
No full textMalate dehydrogenase (MDH) plays a central role in the tricarboxylic acid (TCA) cycle and has recently emerged as a potential drug target due to its involvement in various diseases, including cancer and metabolic disorders. This project investigates the functional impact of a phosphomimetic mutation in the cytosolic isoform of Saccharomyces cerevisiae malate dehydrogenase 2 (ScMDH2), a yeast homolog of human MDH1. By substituting threonine at position 6 with aspartic acid (T6D), we aimed to mimic constitutive phosphorylation at a residue suspected of regulatory modification. The mutant and wildtype enzymes were expressed in E. coli and assayed for kinetic activity by monitoring NADH consumption. The T6D mutant demonstrated a 75% increase in reaction rate compared to wildtype, suggesting improved enzymatic function due to the phosphomimetic substitution. These findings provide insight into the potential regulatory role of phosphorylation in MDH activity. This work also displays the value of course-based undergraduate research in advancing enzyme characterization and drug discovery efforts by implementing this work into a university laboratory course
Tray Carry Assistive Device
No full textThis project involved the design of a cane device that allows users to carry objects whilst using it for support. The goal of this project was to promote the independence of cane users through a compact, lightweight system that maintains full structural stability. Included within the overall system is collapsible storage that assists in carrying common household items ranging in size and shape, including textbooks, coins, or an open cup. The brake system consists of wheels and a retracting rubber foot-stop that allows for travel over doorway barriers and obstacles alike while ensuring smooth travel to protect the contents within the storage, without the need to pick up the overall system. This project will help those with limited mobility feel more independent and live their lives more comfortably
Design of a 3-D Printable Biochar-PET Composite for Remediation of Water Contamination
No full textThe solid product from pyrolysis of biomass - such as fruit waste, dead leaves, or any organic waste - is a carbon-rich material, called biochar, that is widely known for its adsorptive properties, porosity, surface charge, and high surface area. It acts as an effective medium for environmental remediation, such as the removal of organic and inorganic materials from soil and water. Recently, there have been developments in biochar-plastic composites, with biochar acting as a sustainable reinforcing filler, improving the mechanical properties of the plastic. Thus, there is a potential of using these composites in additive manufacturing (3D printing). Additionally, the combination of the biochar’s adsorptive properties combined with the strengthened properties of the plastic could make for an effective material to be applied to the water remediation process. In this project, a biochar-PET (polyethylene terephthalate waste 1) composite filament was developed through extrusion. The biochar was produced by pyrolyzing apple waste from cider production (waste 2). The filament obtained contained 20 wt% biochar, which was decided based on past research papers testing biochar-plastic composites. The composite was designed to be further used as a compatible filament for 3D printing, such as conventional filaments of 1.75 or 2.85mm diameters. To test the water treatment capabilities of the composite, a scalable honeycomb structure was designed with an online CAD software. To allow for scalability, individual parts were printed initially, that would be connected to each other like LEGO® pieces to form the final honeycomb structure. With water flow and high surface area in mind, the shape was designed to have multiple large open areas where water will still contact the material. The biochar-PET filament was extruded successfully but - due to limited equipment - its diameter was inconsistent, and thus, was unfit for 3D printing. However, this problem can be solved in future work with the correct diameter nozzle for the extruder and or an automatic puller system. Consequently, the honeycomb structure was printed with manufactured filament used for 3D printers on campus. Future recommendations include the improvement of the extrusion process to achieve a more consistent filament that can then be used for 3D printing the structure and allowing it to be tested for remediation
PolySolve: A graph-based equation solver combining deterministic computation with machine-learning models for accurate step-by-step solutions.
No full textPolySolve is a graph-based calculator that solves multi-step, domain-specific problems by uniting a deterministic computational core with three narrowly-scoped helpers, one Support Vector Machine and two Large Language Models (LLMs). Existing computational tools, including advanced calculators and Large Language Models (LLMs), face limitations when solving complex, multi-step problems requiring domain-specific reasoning. Traditional calculators are fast and reliable but inflexible, as they require precise input, lack the capability to automatically select and apply equations, and cannot generate contextual natural language explanations. LLMs, on the other hand, are very flexible but can suffer from factual inaccuracies, lack the reliability needed for precise computation, and are computationally expensive for high-parameter models. PolySolve introduces a hybrid methodology that provides the benefits of both approaches. A deterministic core maps equations and variables to a bipartite graph and applies a dependency-guided search algorithm to automatically select relevant equations and derive accurate solution steps. Three small machine learning models are integrated for strictly non-computational tasks: A fine-tuned LLM extracts information from natural language input, such as word problems, into a predefined format, which is then passed to our deterministic core for a solution. An SVM-based variable matcher handles input variations such as typos, abbreviations, or synonyms of known variables, improving robustness at input time. Finally, the output LLM takes the algebraic steps from the deterministic core, and converts them to a natural language explanation with added context. Crucially, none of these models are involved in computing the answer, and only serve to enhance the usability of the calculator. Together, these components deliver accurate, verifiable solutions that end-to-end LLMs cannot guarantee, with the added flexibility and natural language output that traditional calculators lack, closing the gap between computational rigor and user-friendly LLMs
Cardiomyocyte Development through Substrate Design
No full textCardiac patches are a potential treatment option for myocardial infarction caused by cardiovascular diseases. However, methods to obtain mature cardiomyocytes to load these patches are limited, therefore we designed a polyacrylamide hydrogel with varying stiffness and micropatterns to determine their effects on cardiac differentiation. We selected stiffnesses of 3, 8, and 19 kPa, as these fall within the stiffness ranges for embryonic and adult cardiac tissue. We used a raised nodular array to replicate mechanics in utero and an indented linear pattern to promote cell alignment. We observed more cell alignment on higher stiffnesses and embryoid body attachment to micropatterns. However, more research is needed to determine a definitive effect on cardiac differentiation efficiency
And the World Went Dark
No full textAnd the World Went Dark is a “you are what you wear” turn-based roguelite video game that asks the player to balance optimal strategy with the desires of their party members. This game features items that can be equipped to party members to give them unique abilities to use in turn-based combat as the player plays through procedurally-generated runs, losing all gameplay progress when their run ends while preserving narrative progress
FSAE Electric Car - 2024-25
No full textThis year, the Formula Electric MQP has focused on enhancing the electric race car designed last year. We competed in the Formula Hybrid and Electric Competition, an interdisciplinary design and engineering challenge, allowing students to put theory into practice, utilizing kinematic analysis, 3D design, and CNC manufacturing. Our primary objective this year was to develop a reliable and performant vehicle. Our secondary objective was to develop and practice new skills in order to increase our competitive edge in future years. Developing and validating simulation tools with real-world data allowed our team to make educated design decisions and was a key element to achieving this goal. Some notable improvements this year included the design and manufacturing of a new accumulator, a custom steering wheel with advanced driver assistance features, a comprehensive sensor suite that gathered data for vehicle improvement, and our first-ever aerodynamics package
TerraForma
No full textTerraforma is a tactical strategy role playing game (RPG) in which the player uses various magical characters and their associated spells to change the battlefield they fight in. Fires burn the grass beneath a foe's feet, water soaks dirt into becoming mud, and stones erupt from the ground to create unstable terrain. Alongside the game’s main characters, various summoned beasts support the player with abilities that sync with their summoner’s. As the terrain changes, so do the strategies needed to defeat the enemies that stand in their way. The goal of this Major Qualifying Project (MQP) was to develop a complete game concept and produce a polished demo of the game. Said demo took the form of a vertical slice of the game, featuring 3 characters with their own elemental properties and three levels to match. The game was made in Unity’s engine, with the game’s art and sound resources being produced in various other applications. This report documents the development process, focusing on the challenges, changes, and decisions that needed to be made in developing this demo
Planning Solar Power and Water Systems for FiTiCAS Agroecology Projects
No full textPuerto Rico is one of the most import-dependent places in the world, facing food and power shortages from failing infrastructure, outsourced production, and natural disasters. A grassroots initiative founded in 2019 after the destruction of Hurricane Maria, FiTiCAS works to reclaim farmland and restore food sovereignty in Puerto Rico. As this project aimed to design a water and electricity planning guide for the Las Perdices Estate entrusted to FiTiCAS, we analyzed the existing structures on the land, interviewed experts, and researched similar systems. We developed a Utilities Analysis Report, a Utilities Calculator, and a How-to Guide for improving solar microgrid and rainwater harvesting. These tools will help FiTiCAS and similar organizations improve their systems
Microfluidic Model for Tissue Engineered Heart Valves
No full textAortic heart valve disease is highly prevalent and can disrupt flow through the valve. Current treatments require replacements that have limited durability and cannot grow with patients. Tissue engineered heart valves (TEHVs) offer a promising alternative but require host cells to populate the scaffolds. The shear stress environment within heart valves may influence how cells repopulate a TEHV. To determine the effect of complex shear stress patterns caused by flow in the heart valve on cell invasion into the scaffold we developed a microfluidic device made from silicone rubber that uses controlled vacuum pressure to actuate diaphragms to oscillate fluid flow over cells cultured on extracellular matrix scaffolds. Cells are seeded on the matrix, and shear stress is applied via media flow over a cell monolayer. This resulted in a low-cost, high-throughput system to assess cell behavior under physiologically relevant flow conditions