A5009944817- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Single-cell and spatial transcriptomics
- Cancer-related molecular mechanisms research
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- Online Learning and Analytics
- Advanced Graph Neural Networks
- Intelligent Tutoring Systems and Adaptive Learning
- Topic Modeling
- Gene Regulatory Network Analysis
- Cell Image Analysis Techniques
- Genomics and Phylogenetic Studies
- RNA modifications and cancer
- Biomedical Text Mining and Ontologies
- Data Mining Algorithms and Applications
- MicroRNA in disease regulation
- Domain Adaptation and Few-Shot Learning
- RNA Research and Splicing
- Circular RNAs in diseases
- Complex Network Analysis Techniques
- Genomics and Chromatin Dynamics
- Microbial Metabolic Engineering and Bioproduction
- Rough Sets and Fuzzy Logic
- Chromosomal and Genetic Variations
Northwestern Polytechnical University
2016-2025
Ministry of Industry and Information Technology
2018-2024
Beijing Institute of Big Data Research
2023-2024
Northwestern Polytechnic University
2022-2024
University Hospital Magdeburg
2005
Otto-von-Guericke University Magdeburg
2005
Abstract Motivation A microRNA (miRNA) is a type of non-coding RNA, which plays important roles in many biological processes. Lots studies have shown that miRNAs are implicated human diseases, indicating might be potential biomarkers for various types diseases. Therefore, it to reveal the relationships between and diseases/phenotypes. Results We propose novel learning-based framework, MDA-CNN, miRNA-disease association identification. The model first captures interaction features diseases...
Previous Sentiment Analysis (SA) studies have demonstrated that exploring sentiment cues from multiple synchronized modalities can effectively improve the SA results. Unfortunately, until now there is no publicly available dataset for multimodal of stock market. Existing datasets market only provide textual comments, which usually contain words with ambiguous sentiments or even sarcasm expressing opposite literal meaning. To address this issue, we introduce a Fine-grained Multimodal built...
Accurately identifying potential drug-target interactions (DTIs) is a key step in drug discovery. Although many related experimental studies have been carried out for DTIs the past few decades, biological experiment-based DTI identification still timeconsuming and expensive. Therefore, it of great significance to develop effective computational methods DTIs. In this paper, we novel 'end-to-end' learning-based framework based on heterogeneous 'graph' convolutional networks 'DTI' prediction...
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...
Identifying genes associated with Parkinson's disease plays an extremely important role in the diagnosis and treatment of disease. In recent years, based on guilt-by-association hypothesis, many methods have been proposed to predict disease-related genes, but few these are designed or used for gene prediction. this paper, we propose a novel prediction method prediction, named N2A-SVM. N2A-SVM includes three parts: extracting features network, reducing dimension using deep neural predicting...
Abstract Background Drug-target interaction prediction is of great significance for narrowing down the scope candidate medications, and thus a vital step in drug discovery. Because particularity biochemical experiments, development new drugs not only costly, but also time-consuming. Therefore, computational target interactions has become an essential way process discovery, aiming to greatly reducing experimental cost time. Results We propose learning-based method based on feature...
Abstract Motivation The emergence of abundant biological networks, which benefit from the development advanced high-throughput techniques, contributes to describing and modeling complex internal interactions among entities such as genes proteins. Multiple networks provide rich information for inferring function or To extract functional patterns based on multiple heterogeneous network embedding-based methods, aiming capture non-linear low-dimensional feature representation biology, have...
Student performance prediction (SPP) aims to evaluate the grade that a student will reach before enrolling in course or taking an exam. This problem is kernel task toward personalized education and has attracted increasing attention field of artificial intelligence educational data mining (EDM). paper provides systematic review SPP study from perspective machine learning mining. partitions into five stages, i.e., collection, formalization, model, prediction, application. To have intuition on...
The recent advent of CRISPR and other molecular tools enabled the reconstruction cell lineages based on induced DNA mutations promises to solve ones more complex organisms. To date, no lineage algorithms have been rigorously examined for their performance robustness across dataset types number cells. benchmark such methods, we decided organize a DREAM challenge using in vitro experimental intMEMOIR recordings silico data C. elegans tree about 1,000 cells Mus musculus 10,000 Some 22...
Predicting and understanding student learning performance has been a long-standing task in science, which can benefit personalized teaching learning. This study shows that the progress towards this be accelerated by using record data to feed deep model considers intrinsic course association structured features. We proposed multi-source sparse attention convolutional neural network (MsaCNN) predict grades general formulation. MsaCNN adopts multi-scale convolution kernels on grade records...
Sentiment Analysis (SA) is a technique to study people’s attitudes related textual data generated from sources like Twitter. This suggested powerful and effective that can tackle the large contents specifically examine attitudes, sentiments, fake news of “E-learning”, which considered big challenge, as online education sector great importance. On other hand, misinformation COVID-19 have confused parents, students, teachers. An efficient detection approach should be used gather more precise...
Abstract Emerging studies have shown that circular RNAs (circRNAs) are involved in a variety of biological processes and play key role disease diagnosing, treating inferring. Although many methods, including traditional machine learning deep learning, been developed to predict associations between circRNAs diseases, the function has not fully exploited. Some methods explored disease-related based on different views, but how efficiently use multi-view data about circRNA is still well studied....
Cell type identification is an indispensable analytical step in single-cell data analyses. To address the high noise stemming from gene expression data, existing computational methods often overlook biologically meaningful relationships between genes, opting to reduce all genes a unified space. We assume that such can aid characterizing cell features and improving recognition accuracy. this end, we introduce scPriorGraph, dual-channel graph neural network integrates multi-level biosemantics....
Abstract Spatial transcriptomics provides valuable insights into gene expression within the native tissue context, effectively merging molecular data with spatial information to uncover intricate cellular relationships and organizations. In this deciphering domains becomes essential for revealing complex dynamics structures. However, current methods encounter challenges in seamlessly integrating information, resulting less informative representations of spots suboptimal accuracy domain...
Identifying the genes associated to human diseases is crucial for disease diagnosis and drug design. Computational approaches, esp. network-based have been recently developed identify disease-related effectively from existing biomedical networks. Meanwhile, advance in biotechnology enables researchers produce multi-omics data, enriching our understanding on diseases, revealing complex relationships between diseases. However, none of computational approaches able integrate huge amount omics...
Gene Ontology (GO) is one of the most popular bioinformatics resources. In past decade, Ontology-based gene semantic similarity has been effectively used to model gene-to-gene interactions in multiple research areas. However, existing approaches rely only on GO annotations and structure, or incorporate local co-functional network. This may lead inaccurate GO-based resulting from incomplete topology structure annotations.We present NETSIM2, a new network-based method that allows researchers...
The Gene Ontology (GO) is a community-based bioinformatics resource that employs ontologies to represent biological knowledge and describes information about gene product function. GO includes three independent categories: molecular function, process cellular component. For better reasoning, identifying the relationships between terms in different categories are important. However, existing measurements calculate similarity either developed by using data only or take part of combined...
Single cell RNA sequencing (scRNA-seq) is applied to assay the individual transcriptomes of large numbers cells. The gene expression at single-cell level provides an opportunity for better understanding function and new discoveries in biomedical areas. To ensure that based data are interpreted appropriately, it crucial develop computational methods. In this article, we try re-construct a neural network on Gene Ontology (GO) dimension reduction scRNA-seq data. By integrating GO with both...
Arabic is one of the most semantically and syntactically complex languages in world. A key challenging issue text mining summarization, so we propose an unsupervised score-based method which combines vector space model, continuous bag words (CBOW), clustering, a statistically-based method. The problems with multidocument summarization are noisy data, redundancy, diminished readability, sentence incoherency. In this study, adopt preprocessing strategy to solve noise problem use word2vec model...