About

Liuyang Wang is a researcher involved in a project titled 'TNF alpha and Recovery from Alcoholic Liver Injury' at Duke University. The project is funded by the National Institute on Alcohol Abuse & Alcoholism and is administered by the Department of Medicine, Gastroenterology. The research focuses on understanding the role of TNF alpha in the recovery process from alcoholic liver injury. The project started on August 1, 2014, and is scheduled to end on April 30, 2030. Further details about his specific background, academic qualifications, or other contributions are not provided in the page text.

Research topics

• Biology
• Genetics
• Cell biology
• Medicine
• Virology
• Immunology
• Computational biology
• Bioinformatics
• Microbiology
• Cancer research
• Internal medicine

Selected publications

• Decoupling Ego-Motion from Target Dynamics via Dual-Interval Motion Cues for UAV Detection

  ArXiv.org · 2026-05-21

  articleOpen access1st authorCorresponding

  Object detection from Unmanned Aerial Vehicles (UAVs) is challenged by severe ego-motion, camera jitter, and large scale variations. While modern detectors perform well on static images, their direct application to UAV video often fails, particularly for small objects in dynamic scenes. Existing motion-based methods either rely on computationally expensive optical flow or use single-interval differencing, which is sensitive to jitter and limited in capturing diverse motion patterns. We propose a vision-only motion-guided detection framework that decouples target motion from camera-induced disturbances. A homography-based Global Motion Compensation (GMC) first aligns adjacent frames. We then introduce a Dual-Interval Motion Extraction strategy that captures both short-term and long-term motion cues. To integrate these cues, a lightweight Motion-Guided Attention (MGA) module enhances feature representations within a Feature Pyramid Network. Experiments on the VisDrone-VID dataset demonstrate consistent improvements over a strong YOLOv8 baseline under severe ego-motion. Ablation studies further confirm the effectiveness of the dual-interval design and the proposed motion-guided attention mechanism.

  PublisherOA PDF

• Decoupling Ego-Motion from Target Dynamics via Dual-Interval Motion Cues for UAV Detection

  arXiv (Cornell University) · 2026-05-21

  preprintOpen access1st authorCorresponding

  Object detection from Unmanned Aerial Vehicles (UAVs) is challenged by severe ego-motion, camera jitter, and large scale variations. While modern detectors perform well on static images, their direct application to UAV video often fails, particularly for small objects in dynamic scenes. Existing motion-based methods either rely on computationally expensive optical flow or use single-interval differencing, which is sensitive to jitter and limited in capturing diverse motion patterns. We propose a vision-only motion-guided detection framework that decouples target motion from camera-induced disturbances. A homography-based Global Motion Compensation (GMC) first aligns adjacent frames. We then introduce a Dual-Interval Motion Extraction strategy that captures both short-term and long-term motion cues. To integrate these cues, a lightweight Motion-Guided Attention (MGA) module enhances feature representations within a Feature Pyramid Network. Experiments on the VisDrone-VID dataset demonstrate consistent improvements over a strong YOLOv8 baseline under severe ego-motion. Ablation studies further confirm the effectiveness of the dual-interval design and the proposed motion-guided attention mechanism.

  PublisherDOI

• Choline-retinoic acid ionic liquid as a potential therapy agent for acute promyelocytic leukemia

  Journal of Drug Delivery Science and Technology · 2026-01-19

  article
  PublisherDOI

• Identification of Δ9 and Δ11 Desaturases Involved in Sex Pheromone Biosynthesis in <i>Mythimna loreyi</i> (Lepidoptera: Noctuidae)

  Journal of Agricultural and Food Chemistry · 2025-05-01 · 2 citations

  article1st author

  In moths, sex pheromones are synthesized in pheromone glands (PGs) by a variety of enzymes. Desaturases (DESs) are critical for the introduction of double bonds into pheromones. In Mythimna loreyi, the specific DESs involved in sex pheromone biosynthesis remain unclear. In this study, we identified and characterized 25 putative DESs from the M. loreyi genome. Nineteen of them were expressed in the female PGs, with seven showing significant upregulation in response to pheromone biosynthesis activating neuropeptide (PBAN). RNAi-based knockdown of MlorDES2 significantly reduced the titer of pheromone components by 58.6–85.9%, while knockdown of MlorDES9 specifically reduced the production of (Z)-9-tetradecenyl acetate. Functional verification in yeast revealed that MlorDES2 and MlorDES9 exhibited Δ11 and Δ9 desaturase activities, respectively. Taken together, these results collectively demonstrate that MlorDES2 and MlorDES9 are involved in sex pheromone biosynthesis of M. loreyi, suggesting that DESs could be used as potential targets for pest management.

  PublisherDOI

• Multi-source data fusion for estimating potato transpiration under water stress using machine learning models

  Agricultural Water Management · 2025-11-20

  articleOpen access

  Accurate estimation of crop transpiration (T) is critical for optimizing irrigation and enhancing water use efficiency. This study developed a multi-source data fusion framework to estimate daily cumulative transpiration in potato under varying water stress by integrating canopy indices from images with meteorological measurements. The Crop Water Stress Index (CWSI) and Relative Leaf Area Index (RLAI) were extracted using semantic segmentation and image registration. These indices, combined with air temperature, humidity, CO₂ concentration, light intensity measured via a LoRa wireless sensor network, as well as normalized time indicators (hour and minute, scaled to a 0–1 range) corresponding to each observation, served as inputs for Random Forest Regression (RFR), Back-Propagation Neural Network (BPNN), and Long Short-Term Memory (LSTM) models. Six datasets collected over two years (2022 and 2024) with three irrigation treatments were analyzed. Compared to using meteorological variables alone, incorporating CWSI and RLAI significantly enhanced model performance, increasing R² by 1.77–18.44 %, 3.44–11.87 %, and 0.44–18.42 % for RFR, BPNN, and LSTM respectively. In stable environmental conditions of 2022, RFR achieved the best accuracy (R² = 0.8851–0.9654, RMSE = 2.60–9.63 g, MAE = 1.83–6.06 g, RPD = 2.96–5.49). Under more variable conditions in 2024, LSTM outperformed other models (R² = 0.9187–0.9898, RMSE = 14.36–21.02 g, MAE = 10.92–14.62 g, RPD = 3.64–10.54). These findings suggest that RFR is preferable for stable environments, while LSTM is better suited to fluctuating conditions. Integrating CWSI and RLAI with meteorological data improves daily cumulative transpiration estimation, offering a robust foundation for precision irrigation management in potato production.

  PublisherDOI

• Real-Time Object Detection and Motion Estimation on UAVs With Onboard Edge Computing

  2025-07-14

  article1st authorCorresponding

  Computer vision technology has seen widespread application due to advancements in deep learning, yet its implementation on unmanned aerial vehicles (UAVs) is still constrained by hardware limitations and computational power, which pose significant challenges for real-time processing. This paper presents the design of a real-time system deployed on an onboard host computer of a UAV. The system employs a stereo camera to capture both RGB and depth images, facilitating object detection and the estimation of position and speed through YOLO and optical flow algorithms. Furthermore, the UAV’s hardware architecture, including modifications to sensors and the onboard host, is optimized to fulfill the system’s performance requirements. Experiments are conducted in outdoor environments to validate the proposed system’s effectiveness. The results demonstrate that the system achieves a processing speed of 3-6 Hz across the entire pipeline, which includes target detection, optical flow analysis, and position and velocity estimation. As a solution operating on a low-power host, the system exhibits application potential and lays the groundwork for future performance optimization.

  PublisherDOI

• Context-specific eQTLs provide deeper insight into causal genes underlying shared genetic architecture of COVID-19 and idiopathic pulmonary fibrosis

  Human Genetics and Genomics Advances · 2025-01-27 · 3 citations

  articleOpen access

  Most genetic variants identified through genome-wide association studies (GWASs) are suspected to be regulatory in nature, but only a small fraction colocalize with expression quantitative trait loci (eQTLs, variants associated with expression of a gene). Therefore, it is hypothesized but largely untested that integration of disease GWAS with context-specific eQTLs will reveal the underlying genes driving disease associations. We used colocalization and transcriptomic analyses to identify shared genetic variants and likely causal genes associated with critically ill COVID-19 and idiopathic pulmonary fibrosis. We first identified five genome-wide significant variants associated with both diseases. Four of the variants did not demonstrate clear colocalization between GWAS and healthy lung eQTL signals. Instead, two of the four variants colocalized only in cell type- and disease-specific eQTL datasets. These analyses pointed to higher ATP11A expression from the C allele of rs12585036, in monocytes and in lung tissue from primarily smokers, which increased risk of idiopathic pulmonary fibrosis (IPF) and decreased risk of critically ill COVID-19. We also found lower DPP9 expression (and higher methylation at a specific CpG) from the G allele of rs12610495, acting in fibroblasts and in IPF lungs, and increased risk of IPF and critically ill COVID-19. We further found differential expression of the identified causal genes in diseased lungs when compared to non-diseased lungs, specifically in epithelial and immune cell types. These findings highlight the power of integrating GWASs, context-specific eQTLs, and transcriptomics of diseased tissue to harness human genetic variation to identify causal genes and where they function during multiple diseases.

  PublisherOA PDFDOI

• Human genetic variation reveals FCRL3 is a lymphocyte receptor for Yersinia pestis

  Cell Genomics · 2025-06-09 · 2 citations

  articleOpen access

  ). Overexpressed FCRL3 facilitated attachment and invasion of Y. pestis and colocalized with Y. pestis at attachment sites. These properties were variably conserved across the FCRL family, revealing an immunoglobulin-like domain and signaling motifs shared by FCRL3 and FCRL5 to be necessary for attachment and invasion. Direct binding to FCRL5 extracellular domain was confirmed, and B cells (the primary cells that express FCRLs) were preferentially invaded by Y. pestis. Thus, Y. pestis hijacks FCRL proteins, possibly taking advantage of an immune receptor to create a lymphocyte niche during infection.

  PublisherOA PDFDOI

• Amyloid-β precursor protein promotes tumor growth by establishing an immune-exclusive tumor microenvironment

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-02-14 · 1 citations

  preprint

  Abstract During initiation and progression, cancerous tissue hijacks a series of elaborate tissue homeostatic mechanisms to avoid immune surveillance, including neuro-immune interactions. Here, we show that amyloid-β precursor protein (APP) and its β-cleavage product amyloid-β1-42 (Aβ1-42), well-known in the pathogenesis of Alzheimer’s disease (AD), are expressed in multiple cancer tissues. However, the oncogenic activity of APP is due to its E1 domain, instead of Aβ1-42. Mechanistically, APP restricts immune cells influx into tumor microenvironment (TME) and impairs CD8+ T cell and NK cell-based immunity, by dampening type I interferon (IFN) response in TME. We also provide proof-of-concept that vaccination targeting APP is effective for cancer prevention. Our current study reveals a previously unrecognized role of APP in cancer immune surveillance, and provides a new strategy for cancer prevention and treatment by targeting APP.

  PublisherDOI

• Choline-Retinoic Acid Ionic Liquid [Cho][Ra] as Potential Adjuvant to Enhance Humoral, Cellular, and Mucosal Immune Responses of SARS-CoV-2 RBD Antigen

  Molecular Pharmaceutics · 2025-09-17 · 1 citations

  article

  All-trans retinoic acid (Ra) has been demonstrated to enhance the establishment of systemic and intestinal immunity for codelivered antigens; however, the extremely poor water solubility of Ra significantly limits its application in vaccines. Herein, leveraging the advantages of ionic liquids in modifying the physicochemical properties of small-molecule drugs, we synthesized a novel ionic liquid composed of choline and retinoic acid ([Cho][Ra]). Compared to Ra, [Cho][Ra] exhibited an 18165-fold increase in solubility at pH 7.4 and an acid-responsive Ra release behavior. The [Cho][Ra] was then formulated with the SARS-CoV-2 RBD antigen, and its adjuvant effects were comprehensively evaluated. The in vitro cellular studies indicated that at 50 μg Ra equivalent dose, the [Cho][Ra]@RBD showed significantly higher than RBD along in activation of BMDC and the expression of gut homing molecules, including C–C motif chemokine receptor 9 (CCR9, 10.55 folds), α4β7 integrin (13.29 folds), and lymph node migration molecule CCR7 (82.68 folds). Animal experiments showed that compared to RBD alone, [Cho][Ra]@RBD promoted the establishment of mucosal immunity in the intestines in mice at the early stage following subcutaneous immunization and elicited higher serum antibody levels. Furthermore, compared to RBD and Alum@RBD, [Cho][Ra]@RBD enhanced the formation of CD4+ TCM and CD8+ TCM cells, indicating stronger cellular immune responses. In summary, [Cho][Ra] showed potential adjuvant effects in enhancing the humoral and cellular immune responses, as well as in the establishment of intestinal mucosal immunity for RBD antigen without the need for delivery system, thereby offering promising prospects for applying Ra in vaccines.

  PublisherDOI

Recent grants

• AN ATLAS OF SHARED GENETIC ARCHITECTURE CONNECTING CELL BIOLOGY AND HUMAN DISEASE

  NIH · $440k · 2018–2021

Frequent coauthors

• Dennis C. Ko

  Duke University

  71 shared

• Qi-Jing Li

  Agency for Science, Technology and Research

  29 shared

• Jianquan Liu

  Lanzhou University

  27 shared

• Xiangdong Mei

  24 shared

• Dongmei She

  Chinese Academy of Agricultural Sciences

  19 shared

• Yujin Wang

  NGM Biopharmaceuticals (United States)

  19 shared

• Christopher W. Woods

  Duke Medical Center

  18 shared

• Alejandro L. Antonia

  Duke Medical Center

  17 shared

Labs

• Ko LabPI