http://hdl.handle.net/10453/35199 2026-06-06T16:34:49Z http://hdl.handle.net/10453/195238 Title: Prompt-based multimodal representation learning for drug repurposing. Authors: Liu, J; U, K; Rana, D; Meixuan Zhang, S; Yu, J; Yang, S; Jin, B; Wang, X; Yang, Z; Tang, H; Zhao, J Abstract: Drug repurposing significantly reduces development costs and shortens research cycles, making it a critical strategy in drug discovery. An emerging class of drug repurposing approaches applies deep learning to structural data. However, these methods often depend on static representations of molecular and protein structures, which may not fully capture the dynamic character of compound-protein interactions. To address these challenges and enhance the accuracy of compound-protein interaction predictions, we introduce an innovative prompt-based multimodal representation learning framework that dynamically encodes task-specific contextual information for drug repurposing. Specifically, the framework includes a dynamic prompt generation module that adaptively creates receptor-specific prompts and a prompt calibration module for effective multimodal feature integration and optimization. When applied to identifying FDA-approved drug candidates targeting G-protein-coupled receptors, our method achieved a 7.4% improvement in mean absolute error compared with state-of-the-art methods, with up to a 25.1% improvement for specific target-of-interest. By demonstrating potential in repurposing non-opioid treatments without the risk of addiction for safe pain management, our method has the capacity to advance drug discovery and meet a wide range of therapeutic needs. 2025-11-01T00:00:00Z http://hdl.handle.net/10453/195234 Title: Tracking Academic Topic Evolution From the Perspective of Author Interest Shifts Authors: Jiang, W; Zhang, Y; Wang, M; Mo, H; Hussain, O; Dong, D; Zhang, W Abstract: Understanding how academic topics evolve over time is essential for tracking scientific progress and identifying emerging trends This study proposes a new framework for analyzing topic evolution based on the shifting interests of individual authors whose evolving research choices play a key role in shaping topic trajectories By leveraging a time sequenced learning approach that incorporates author information the framework captures how topics emerge grow decline and reemerge Improving traditional topic modeling approaches that rely solely on textual patterns our framework explicitly models the temporal dynamics of individual authors topic engagement providing a micro level lens on macro level topic changes As a case study we apply this framework to the Scientometrics field using data from 1978 to 2024 and construct a topic evolutionary map that reveals strong interconnections among bibliometrics research evaluation and related methodologies The analysis also reveals that despite the persistence of foundational topics like citation analysis and knowledge management which are reinforced by sustained author interest only two genuinely emerging topics were identified suggesting that while author interests evolve they tend to favor established areas over the exploration of entirely new directions 2025-12-22T00:00:00Z http://hdl.handle.net/10453/195228 Title: A Dual Linearly-Polarized Active Cross-Dipole Antenna for Radio Astronomy Authors: Ansari, M; Dunning, A; Bannister, K; Chung, Y; Pathikulangara, J Abstract: This paper presents a dual-linearly polarized active antenna with a wide field of view operating from 550 MHz to 750 MHz. The antenna consists of a PCB crossed dipole suspended above an aluminum cavity to allow for a symmetrical beam with low back radiation. This antenna has built-in low noise amplifier (LNA) with a noise temperature of less than 20K. The antenna exhibits a wide field of view but low gain towards the horizon, which makes it well suited for use in an all-sky monitor aperture array for radio astronomy. 2025-01-01T00:00:00Z http://hdl.handle.net/10453/195224 Title: Techno-Economic Analysis of Design and Transport Parameters in PEMWEs Authors: Bayat, A; Das, PK; Eager, D; Saha, SC Abstract: Green hydrogen has emerged as a promising pathway toward decarbonization of future energy systems, with proton exchange membrane water electrolyzers (PEMWEs) attracting increasing attention due to their high efficiency, compact design, and compatibility with renewable energy sources. This study extends prior multi-physics investigations of proton exchange membrane water electrolyzers (PEMWEs) by translating performance-driven design insights into techno-economic implications for green hydrogen production. Building upon previously developed numerical models that examined the effects of membrane thickness, membrane conductivity, and operating temperature, as well as porosity distributions within porous transport layers and the influence of gas crossover under varying outlet pressures, the present work establishes a direct link between electrochemical behaviour and hydrogen production cost. Simulation-derived polarization characteristics are integrated into a simplified techno-economic framework to quantify variations in energy consummption, hydrogen yield, and levelized cost of hydrogen (LCOH) under different design and operational configurations. Unlike conventional assessments relying on assumed efficiencies or generic performance data, this study employs physics-based simulation outputs as the primary input for economic evaluation, enabling a more faithful representation of design-dependent cost behaviour. The proposed framework further enables direct assessment of how membrane properties, transport characteristics, and structural configurations influence techno-economic performance. The results reveal how subtle changes in transport and structural parameters propagate into measurable economic consequences, highlighting critical trade-offs between efficiency enhancement and cost escalation. Specifically, the investigated design and transport variations resulted in specific energy consumption values ranging from ∼35–55 kWh kg⁻¹ H₂, stack electrical efficiencies of ∼0.5–0.9, and LCOH values of ∼2.3–5 USD kg⁻¹ H₂, demonstrating that relatively small transport-induced performance changes can propagate into measurable economic consequences, particularly at moderate-to-high operating current densities. The findings provide design-oriented economic insights that support informed decision-making for cost-sensitive optimization of PEMWE systems, bridging the gap between electrochemical modelling and real-world deployment considerations. 2026-06-01T00:00:00Z