A5100682453- Single-cell and spatial transcriptomics
- RNA and protein synthesis mechanisms
- RNA modifications and cancer
- Endoplasmic Reticulum Stress and Disease
- Domain Adaptation and Few-Shot Learning
- Cancer-related molecular mechanisms research
- Cellular transport and secretion
- Cardiovascular and Diving-Related Complications
- Epigenetics and DNA Methylation
- DNA and Nucleic Acid Chemistry
- Infective Endocarditis Diagnosis and Management
- RNA Research and Splicing
- Congenital Heart Disease Studies
- Evolution and Genetic Dynamics
- COVID-19 diagnosis using AI
- Cancer-related Molecular Pathways
- Retinal Imaging and Analysis
- Genomics, phytochemicals, and oxidative stress
- Retinal Diseases and Treatments
- Fungal and yeast genetics research
- Non-Invasive Vital Sign Monitoring
- Neonatal and fetal brain pathology
- Tracheal and airway disorders
- Medical Imaging and Analysis
- Advanced biosensing and bioanalysis techniques
Guiyang Medical University
2021-2024
Sun Yat-sen University
2020-2024
Wuhan City Chinese Medicine Hospital
2023
State Key Laboratory of Microbial Technology
2017
Shandong University
2017
Cellulase expression via extracellular secretion or surface display in Saccharomyces cerevisiae is one of the most frequently used strategies for a consolidated bioprocess (CBP) cellulosic ethanol production. However, inefficiency yeast secretory pathway often results low production heterologous proteins, which largely limits cellulase display. In this study, components vesicle trafficking from endoplasmic reticulum (ER) to Golgi and plasma membrane, involved budding, tethering fusion, were...
Accurate identification of spatial domains is essential in the analysis transcriptomics data order to elucidate tissue microenvironments and biological functions. However, existing methods only perform domain segmentation based on local or global relationships between spots, resulting an underutilization information. To this end, we propose SECE, a deep learning-based method that captures both among spots aggregates their information using expression similarity similarity. We benchmarked...
Abstract Motivation Accurate identification of spatial domains is essential for analyzing transcriptomics data to elucidate tissue microenvironments and biological functions. Existing methods utilize either local or global relationships between spots aid domain segmentation. A method that can concurrently capture both information may improve domains. Results In this article, we propose SECE, a deep learning-based captures among aggregates their using expression similarity similarity. We...
<title>Abstract</title> Messenger RNA (mRNA) therapeutics show immense promise, but their efficacy is limited by suboptimal protein expression. Here, we present RiboCode, a deep learning framework that generates mRNA codon sequences for enhanced production. RiboCode introduces several advances, including direct from large-scale ribosome profiling data, context-aware optimization and generative exploration of large sequence space. <italic>In silico</italic> analysis demonstrate RiboCode’s...
Accurate evaluation of retinopathy prematurity (ROP) severity is vital for screening and proper treatment. Current deep-learning-based automated AI systems assessing ROP do not follow clinical guidelines are opaque. The aim this study to develop an interpretable system by mimicking the process determine level. A total 6100 RetCam Ⅲ wide-field digital retinal images were collected from Guangdong Women Children Hospital at Panyu (PY) Zhongshan Ophthalmic Center (ZOC). 3330 520 pediatric...
The hospital admitted a 3-day-old female infant presenting with persistent facial cyanosis and hypoxic symptoms, echocardiography revealed congenitally unguarded tricuspid valve orifice an atrial septal defect. After being followed up until the age of one half years, child underwent bidirectional Glenn's surgery achieved successful survival.
Understanding the basis of DNA-binding specificity transcription factors (TFs) has been long-standing interest. Despite extensive efforts to map millions putative TF binding sequences, identifying critical determinants for DNA remains a major challenge. The coevolution residues in proteins occurs due shared evolutionary history. However, it is unclear how coevolving TFs contribute specificity. Here, we systematically collected publicly available data sets from multiple large-scale...
Diabetic retinopathy (DR) severity grade depends on lesion types. Automatic segmentation of DR fundus image plays a key role in the diagnosis DR. It is increasingly common that model trained by images from different sources. While source domain transferred to another (target domain), performance generally decreases. In this paper, novel method was proposed for cross-domain lesions applying cycle-consistent adversarial networks (CycleGAN) and an improved Xception-based UNet named AttXUNet. To...
A 37-year-old male patient was admitted to the hospital with repeated fever for 15 days and right limb weakness 7 + hours. Brain magnetic resonance imaging diffusion-weighted showed multiple high signal intensities in bilateral cerebral hemisphere, consistent performance of infarction (Panels B). Transthoracic echocardiography revealed an ovoid, hypoechoic, mobile mass (19 mm × 17 mm) (green arrow) left atrium attached root anterior leaflet mitral valve no significant valvular abnormalities,...