Browsing by Author "Lu, Yi"

Now showing 1 - 4 of 4
  • Thumbnail Image
    DevCoach: Supporting Students in Learning the Software Development Life Cycle at Scale with Generative Agents
    Wang, Tianjia; Ramanujan, Ramaraja; Lu, Yi; Mao, Chenyu; Chen, Yan; Brown, Chris (ACM, 2024-07-09)
    Supporting novice computer science students in learning the software development life cycle (SDLC) at scale is vital for ensuring the quality of future software systems. However, this presents unique challenges, including the need for effective interactive collaboration and access to diverse skill sets of members in the software development team. To address these problems, we present “DevCoach”, an online system designed to support students learning the SDLC at scale by interacting with generative agents powered by large language models simulating members with different roles in a software development team. Our preliminary user study results reveal that DevCoach improves the experiences and outcomes for students, with regard to learning concepts in SDLC’s “Plan and Design” and “Develop” phases.We aim to use our findings to enhance DevCoach to support the entire SDLC workflow by incorporating additional simulated roles and enabling students to choose their project topics. Future studies will be conducted in an online Software Engineering class at our institution, aiming to explore and inspire the development of intelligent systems that provide comprehensive SDLC learning experiences to students at scale.
  • Thumbnail Image
    Development of Virus-like particles (VLPs) Based Vaccines Against Porcine Reproductive and  Respiratory Syndrome Virus (PRRSV) and Porcine Epidemic Diarrhea Virus (PEDV)
    Lu, Yi (Virginia Tech, 2020-03-16)
    Porcine reproductive and respiratory syndrome virus (PRRSV) and porcine epidemic diarrhea virus (PEDV) are two of the most prevalent swine pathogens that have impacted the global swine industry for decades. Both are RNA viruses with increasing heterogeneity over the years, making a vaccine solution ever so challenging. Modified live-attenuated vaccines (MLVs) have been the most common approach, but the long-term safety regarding their potential for pathogenic reversion still needs to be addressed. Subunit based vaccines have been the focus of numerous development studies around the world with renewed interest in their promising prospects in both safety and efficacy. Our lab has developed a unique approach to use hepatitis B virus core capsid protein (HBcAg) as a vaccine delivery vehicle for either PRRSV or PEDV viral epitope antigens. Recombinantly produced HBcAg forms an icosahedral capsid virus-like particle (VLP) that has 240 repeats in a single assembled particle. By inserting different epitope antigens from these porcine pathogens into the particle, we can achieve repetitive antigen presentation to the host's immune system by taking advantage of the polymeric nature of VLP. The first animal study evaluated the efficacy of 4 VLP based vaccine candidates against PRRSV in mice. These 4 vaccines incorporated 2 B-cell epitopes (61QAAIEVYEPGRS72 and 89ELGFVVPPGLSS100) and 2 T-cell epitopes (117LAALICFVIRLAKNC131 and 149KGRLYRWRSPVIIEK163) from PRRSV structural proteins GP3 and GP5 respectively. Candidate GP3-4 was able to stimulate a significant viral neutralizing response in mouse sera against two PRRSV strains, one being heterologous, demonstrating its potential of cross-protection against PRRSV. The second animal study took an optimized VLP vaccine candidate against PEDV from previous development studies in mice, and assessed its efficacy through a comprehensive pregnant gilt vaccination and neonatal piglet challenge model. The vaccine candidate incorporated B-cell epitope 748YSNIGVCK755 from the PEDV spike protein. It was able to elicit significant viral neutralization antibody titer in gilt milk at 3 days post-farrowing (DPF), and provided nursing piglets with clinical relief in terms of morbidity, viral shedding, small intestinal lesions, and 10 days post-challenge (DPC) survival rate.
  • Thumbnail Image
    Helping job seekers prepare for technical interviews by enabling context-rich interview feedback
    Lu, Yi (Virginia Tech, 2024-06-11)
    Technical interviews have become a popular method for recruiters in the tech industry to assess job candidates' proficiency in both soft skills and technical skills as programmers. However, these interviews can be stressful and frustrating for interviewees. One significant cause of the negative experience of technical interviews was the lack of feedback, making it difficult for job seekers to improve their performance progressively by participating in technical interviews. Although there are open platforms like Leetcode that allow job seekers to practice their technical proficiency, resources for conducting mock interviews to practice soft skills like communication are limited and costly to interviewees. To address this, we investigated how professional interviewers provide feedback if they were conducting a mock interview and the difficulties they face when interviewing job seekers by running mock interviews between software engineers and job seekers. With the insights from the formative studies, we developed a new system for technical interviews aiming to help interviewers conduct technical interviews with less cognitive load and provide context-rich feedback. An evaluation study on the usability of using our system to conduct technical interviews further revealed the unresolved cognitive loads of interviewers, underscoring the requirements for further improvement to facilitate easier interview processes and enable peer-to-peer interview practices.
  • Thumbnail Image
    Time-Course Transcriptome Profiling Reveals Differential Resistance Responses of Tomato to a Phytotoxic Effector of the Pathogenic Oomycete Phytophthora cactorum
    Zhou, Xue; Wen, Ke; Huang, Shen-Xin; Lu, Yi; Liu, Yang; Jin, Jing-Hao; Kale, Shiv D.; Chen, Xiao-Ren (MDPI, 2023-02-15)
    Blight caused by Phytophthora pathogens has a devastating impact on crop production. Phytophthora species secrete an array of effectors, such as Phytophthora cactorum-Fragaria (PcF)/small cysteine-rich (SCR) phytotoxic proteins, to facilitate their infections. Understanding host responses to such proteins is essential to developing next-generation crop resistance. Our previous work identified a small, 8.1 kDa protein, SCR96, as an important virulence factor in Phytophthora cactorum. Host responses to SCR96 remain obscure. Here, we analyzed the effect of SCR96 on the resistance of tomato treated with this recombinant protein purified from yeast cells. A temporal transcriptome analysis of tomato leaves infiltrated with 500 nM SCR96 for 0, 3, 6, and 12 h was performed using RNA-Seq. In total, 36,779 genes, including 2704 novel ones, were detected, of which 32,640 (88.7%) were annotated. As a whole, 5929 non-redundant genes were found to be significantly co-upregulated in SCR96-treated leaves (3, 6, 12 h) compared to the control (0 h). The combination of annotation, enrichment, and clustering analyses showed significant changes in expression beginning at 3 h after treatment in genes associated with defense and metabolism pathways, as well as temporal transcriptional accumulation patterns. Noticeably, the expression levels of resistance-related genes encoding receptor-like kinases/proteins, resistance proteins, mitogen-activated protein kinases (MAPKs), transcription factors, pathogenesis-related proteins, and transport proteins were significantly affected by SCR96. Quantitative reverse transcription PCR (qRT-PCR) validated the transcript changes in the 12 selected genes. Our analysis provides novel information that can help delineate the molecular mechanism and components of plant responses to effectors, which will be useful for the development of resistant crops.