A5067965165- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Molecular Biology Techniques and Applications
- Topic Modeling
- Cancer-related molecular mechanisms research
- Cloud Computing and Resource Management
- Immunotherapy and Immune Responses
- Energy Load and Power Forecasting
- Bayesian Modeling and Causal Inference
- Speech and Audio Processing
- Advanced Algorithms and Applications
- RNA modifications and cancer
- Complex Network Analysis Techniques
- Computational Drug Discovery Methods
- Rough Sets and Fuzzy Logic
- IoT and Edge/Fog Computing
- Machine Learning in Bioinformatics
- Autophagy in Disease and Therapy
- Target Tracking and Data Fusion in Sensor Networks
- Data Mining Algorithms and Applications
- AI in cancer detection
- Underwater Acoustics Research
- Microbial Metabolic Engineering and Bioproduction
- Advanced Computational Techniques and Applications
- RNA Research and Splicing
Lanzhou University
2024
Chinese University of Hong Kong
2024
Chongqing University of Posts and Telecommunications
2018-2023
National University of Defense Technology
2023
China Medical University
2022
University of Victoria
2021
Northwestern Polytechnical University
2012-2020
PLA Academy of Military Science
2020
Huazhong Agricultural University
2018-2019
Dalian University of Technology
2019
The epithelial-to-mesenchymal transition (EMT) is an essential biological process during embryonic development that also implicated in cancer metastasis. While the transcriptional regulation of EMT has been well studied, role alternative splicing (AS) remains relatively uncharacterized. We previously showed epithelial cell-type-specific proteins regulatory 1 (ESRP1) and ESRP2 are important for many AS events altered EMT. However, contributions ESRPs other regulators to network require...
The classification of cancer subtypes is great importance to disease diagnosis and therapy. Many supervised learning approaches have been applied subtype in the past few years, especially deep based approaches. Recently, forest model has proposed as an alternative neural networks learn hyper-representations by using cascade ensemble decision trees. It proved that competitive or even better performance than some extent. However, standard may face overfitting diversity challenges when dealing...
We consider the auto-scaling problem for application hosting in a cloud, where applications are elastic and number of requests changes over time. The serviced by Virtual Machines (VMs), which reside on Physical (PMs) cloud. aim to minimize PMs intelligently packing VMs into PMs, while auto-scaled, i.e., dynamically acquired released, accommodate varying needs. shadow routing based approach this problem. proposed algorithm employs specially constructed virtual queueing system produce an...
Mining epistatic loci which affects specific phenotypic traits is an important research issue in the field of biology. Bayesian network (BN) a graphical model can express relationship between genetic and phenotype. Until now, it has been widely used into epistasis mining many work. However, this method two disadvantages: low learning efficiency easy to fall local optimum. Genetic algorithm excellence rapid global search avoiding falling It scalable integrate with other algorithms. This work...
Abstract Transcription factor E2F1 has been largely studied as a promoter of S-phase transition in the cell cycle and regulator apoptosis. Recently, shown to regulate wide range genes response inflammatory stimulation macrophages contribute T activation pathogens, implicating an extensive immunological role for E2F1. Dendritic cells (DCs) play critical roles professional APCs development immune responses. However, it is unclear whether any effect on DC phenotype or function. In this paper,...
Abstract Tumour-induced dendritic cell (DC) dysfunction plays an important role in cancer immune escape. However, the underlying mechanisms are not yet fully understood, reflecting lack of appropriate experimental models both vivo and vitro . In present study, study model for tumour-induced DC was established by culturing DCs with pooled sera from multiple non-small lung (NSCLC) patients. The results demonstrated that human monocyte-derived exhibited systematic functional deficiencies....
Identifying cancer genes is vital for diagnosis and treatment. However, because of the complexity occurrence limited knowledge, it hard to identify accurately using only a few omics data, overall performance existing methods being called further improvement. Here, we introduce two-stage gradual-learning strategy GLIMS predict integrative features from multi-omics data. Firstly, uses semi-supervised hierarchical graph neural network initial candidate by integrating data protein-protein...
We consider a shadow routing based approach to the problem of real-time adaptive placement virtual machines (VM) in large data centers (DC) within network cloud. Such particular has respect vector packing constraints on allocation VMs host physical (PM) DC, because each PM can potentially serve multiple simultaneously. Shadow is attractive that it allows variety system objectives and/or be treated common framework (as long as underlying optimization convex). Perhaps even more feature...
The classification of cancer subtypes is great importance in disease diagnosis and therapy. Many supervised learning methods have been applied to the past few years, especially deep based methods. Recently, a forest model has proposed as an alternative neural networks learn hyper-representations by using cascade ensemble decision trees, it proved that competitive or even better performance than networks. However, original may face under-fitting diversity problems when dealing with small...
The identification of cancer subtypes is crucial to diagnosis and treatments. A number methods have been proposed identify by integrating multi-omics data in recent years. However, the existing rarely consider biases similarity between samples weights different omics integration. More accurate flexible integration approaches need be developed comprehensively investigate subtypes. In this paper, we propose a simple fusion model for We each predict corrected similarities using generalized...
Reactive oxygen species (ROS) act as a signaling intermediate to promote cellular adaptation maintain homeostasis by regulating autophagy during pathophysiological stress. However, the mechanism which ROS promotes is still largely unknown. Here, we show that ATM/CHK2/ULK1 axis initiates sensing under metabolic We report ULK1 physiological substrate of CHK2, and binding CHK2 depends on signal phosphorylation threonine 68 Further, phosphorylates serine 556, this dependent ATM/CHK2 pathway....
Abstract The identification of neoantigens is crucial for advancing vaccines, diagnostics, and immunotherapies. Despite this importance, a fundamental question remains: how to model the presentation by major histocompatibility complex class I molecules recognition peptide-MHC-I (pMHC-I) T cell receptors (TCRs). Accurate prediction pMHC-I binding TCR remains significant computational challenge in immunology due intricate motifs long-tail distribution known pairs public databases. Here, we...
Identification of cancer subtypes is great importance to facilitate diagnosis and therapy. A number methods have been proposed integrate multi-sources data identify in recent years. However, few them consider the regulatory associations between genome features contribution weights different data-views integration. It widely accepted that play important roles subtype studies. In addition, may contributions integration for prediction. this paper, we propose a method, CSPRV, improve prediction...
The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is responsible for millions of deaths around the world. To help contribute to understanding crucial knowledge and further generate new hypotheses relevant SARS-CoV-2 human protein interactions, we make use information abundant Biomine probabilistic database extend experimentally identified SARS-CoV-2-human protein-protein interaction (PPI) network in silico. We an extended by integrating from database, PPI other validated...
Abstract Background Single-cell RNA sequencing (scRNA-seq) provides an effective tool to investigate the transcriptomic characteristics at single-cell resolution. Due low amounts of transcripts in single cells and technical biases experiments, raw scRNA-seq data usually includes large noise makes downstream analyses complicated. Although many methods have been proposed impute noisy recent years, few them take into account prior associations across genes imputation integrate multiple types...
We develop shadow routing based online algorithms for the joint problem of application-to-VM and VM-to-PM assignments in a cloud environment. The asymptotic optimality algorithm is proved performance evaluated by simulations.
Mining the protein complexes and functional modules from protein-protein interaction (PPI) networks is vital to understand mechanism of cellular components functions. Most proposed methods had solely focused on static properties PPI since available data are static. However, systems highly dynamic. That is, interactions proteins responsive environmental cues accomplish diverse It important consider dynamic inherent within identify modules. In addition, most computational did not distinguish...