A5100318311- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Single-cell and spatial transcriptomics
- Gene Regulatory Network Analysis
- Advanced Graph Neural Networks
- Computational Drug Discovery Methods
- Machine Learning in Bioinformatics
- Complex Network Analysis Techniques
- Advanced Proteomics Techniques and Applications
- Genomics and Chromatin Dynamics
- Ferroptosis and cancer prognosis
- Data Mining Algorithms and Applications
- Image and Signal Denoising Methods
- Cell Image Analysis Techniques
- Advanced Image Processing Techniques
- Nuclear Receptors and Signaling
- RNA Interference and Gene Delivery
- Sentiment Analysis and Opinion Mining
- Image and Video Quality Assessment
- Glaucoma and retinal disorders
- Viral Infectious Diseases and Gene Expression in Insects
- Cancer, Lipids, and Metabolism
- Emotion and Mood Recognition
- CRISPR and Genetic Engineering
- RNA Research and Splicing
Qingdao Women and Children's Hospital
2024
Qingdao University
2024
Soochow University
2024
Nankai University
2019-2024
Hisense (China)
2024
Tencent (China)
2023
Northeastern University
2022
Shanghai Changzheng Hospital
2020
Harbin Medical University
2011-2015
Shanghai Center For Bioinformation Technology
2015
Gene expression profiling has been widely used to characterize cell status reflect the health of body, diagnose genetic diseases, etc. In recent years, although cost genome-wide is gradually decreasing, collecting profiles for thousands genes still very high. Considering gene expressions are usually highly correlated in humans, values remaining target can be predicted by analyzing 943 landmark genes. Hence, we designed an algorithm predicting based on XGBoost, which integrates multiple tree...
One of the challenging problems in etiology diseases is to explore relationships between initiation and progression abnormalities local regions metabolic pathways. To gain insight into such relationships, we applied "k-clique" subpathway identification method all disease-related gene sets. For each disease, disease risk pathways were then identified considered as subpathways associated with disease. We finally built a disease-metabolic network (DMSPN). Through analyses based on biology,...
Abstract Proteins are crucial for life, and measuring their abundance at the single-cell level can facilitate a high-resolution understanding of biological mechanisms in cellular processes disease progression. However, current proteomic technologies face challenges such as limited coverage, throughput, sensitivity, well batch effects, high costs, stringent experimental operations. Drawing inspiration from translation procedure both natural language processing (NLP) genetic central dogma, we...
Real-world social networks are often hierarchical, re- flecting the fact that some communities composed of a few smaller, sub-communities. This paper describes hierarchical Bayesian model based scheme, namely HSN- PAM (Hierarchical Social Network-Pachinko Allocation Model), for discovering probabilistic, com- munities in networks. scheme is powered by previously developed model. In this classified into two categories: super-communities and regular-communities. Two differ- ent network...
Over the past decades, studies have reported that combinatorial regulation of transcription factors (TFs) and microRNAs (miRNAs) is essential for appropriate execution biological events developmental processes. Dysregulations these regulators often cause diseases. However, there are no available resources on regulatory cascades TFs miRNAs in context human To fulfill this vacancy, we established TMREC database study. First, integrated curated transcriptional post-transcriptional regulations...
Background Traditionally top-down method was used to identify prognostic features in cancer research. That is say, differentially expressed genes usually versus normal were identified see if they possess survival prediction power. The problem that from one set of patient samples can rarely be transferred other datasets. We apply bottom-up approach this study: correlated or clinical stage selected first and prioritized by their network topology additionally, then a small as signature. Methods...
Abstract Background Atherosclerosis (AS) is a pathology factor for cardiovascular diseases and instability of atherosclerotic plaques contributes to acute coronary events. This study identified hub gene VCL discovered its potential therapeutic targets plaques. Methods Differential expressed genes (DEGs) were screened between unstable stable from GSE120521 dataset then used construction protein-protein interactions (PPI) network. Through topological analysis, within this PPI network, followed...
Abstract In recent years, a number of computational approaches have been proposed to effectively integrate multiple heterogeneous biological networks, and shown impressive performance for inferring gene function. However, the previous methods do not fully represent critical neighborhood relationship between genes during feature learning process. Furthermore, it is difficult accurately estimate contributions different views multi-view integration. this paper, we propose MGEGFP, graph...
Technological advances have now made it possible to simultaneously profile the changes of epigenomic, transcriptomic and proteomic at single cell level, allowing a more unified view cellular phenotypes heterogeneities. However, current computational tools for single-cell multi-omics data integration are mainly tailored bi-modality data, so new urgently needed integrate tri-modality with complex associations. To this end, we develop scMHNN based on hypergraph neural network. After modeling...
Intracellular delivery of proteins has attracted significant interest in biological research and cancer treatment, yet it continues to face challenges due the lack effective approaches. Herein, we developed an efficient strategy via cationic α-helical polypeptide-mediated anionic proprotein delivery. The protein was reversibly modified with adenosine triphosphate dynamic covalent chemistry prepare (A-protein) abundant phosphate groups. A guanidyl-decorated polypeptide (LPP) employed not only...
Convolutional neural network (CNN) has been successfully used for the identification of motif occupancy. However, CNN architecture requires varying length instead fixed-length filters due to different lengths. Moreover, plain networks with single point estimation weights suffer from over-fitting, which is more likely occur as increasing parameters multi-scale modeling.Hence, we have designed a Bayesian Multi-scale CNN. The model employs convolutional scales extract latent features DNA...
In protein-protein interaction networks, proteins combine into macromolecular complexes to execute essential functions in the cells, such as replication, transcription, protein transport. Considering certain rate of false positive and negative interactions, we take a confidence probability on interactions correlate gene expression data assign weights edges PPI networks. Then propose CIGE algorithm detecting from Our takes maximal full-connected sub-graph core graph seed node, decides whether...
Molecular signaling events regulate cellular activity. Cancer stimulating signals trigger responses that evade the regulatory control of cell development. To understand mechanism regulation in cancer, it is necessary to identify activated pathways cancer. We have developed RepairPATH, a computational algorithm explores The RepairPATH integrates RepairNET, an assembled protein interaction network associated with DNA damage response, gene expression profiles derived from microarray data. Based...
The recent development of high-throughput biological techniques for functional genomics have generated a large quantity new network data. Analyzing these networks provides novel insights in understanding basic mechanisms controlling cellular processes. In this paper, we integrate protein interaction and microarray data transform the un-weighted protein-protein to its weighted correspondent. We then present graph mining problem, associated patterns across genome-wide network. central idea...
Abstract Background Hepatocellular carcinoma (HCC) is a malignancy causing highly death rate in the world. Despite development of treatment strategies for HCC, prognosis this remains unsatisfactory. In study, we aimed to identify target genes associated with HCC patients. Methods Three expression profiles tissues were extracted from Gene Expression Omnibus database explore differentially expressed (DEGs) using GEO2R method. Functional enrichment analysis was performed reveal biological...