A5100368759- RNA and protein synthesis mechanisms
- Machine Learning in Bioinformatics
- Genomics and Phylogenetic Studies
- Protein Structure and Dynamics
- Network Security and Intrusion Detection
- Biomedical Text Mining and Ontologies
- Remote-Sensing Image Classification
- Topic Modeling
- Advanced Malware Detection Techniques
- Remote Sensing and Land Use
- Bioinformatics and Genomic Networks
- Machine Learning in Healthcare
- Sperm and Testicular Function
- Enzyme Structure and Function
- DNA and Nucleic Acid Chemistry
- Genetic Neurodegenerative Diseases
- Reproductive Physiology in Livestock
- Genetic and phenotypic traits in livestock
- Monoclonal and Polyclonal Antibodies Research
- Genetic diversity and population structure
- Internet Traffic Analysis and Secure E-voting
- Vitamin C and Antioxidants Research
- Genetic Associations and Epidemiology
- Microbial Community Ecology and Physiology
- Neurogenetic and Muscular Disorders Research
China Agricultural University
2025
Changsha Medical University
2025
Shanghai East Hospital
2025
Guizhou University
2023-2024
Huzhou University
2009-2024
Beijing University of Chinese Medicine
2018-2024
Florida State University
2005-2023
Xiangtan University
2023
Ocean University of China
2022
North Carolina State University
2013-2021
We propose a machine learning framework to capture the dynamics of high-frequency limit order books in financial equity markets and automate real-time prediction metrics such as mid-price movement price spread crossing. By characterizing each entry book with vector attributes volume at different levels, proposed builds model for metric help multi-class support machines. Experiments real data establish that features selected by are effective short-term forecasts.
The desert locust (Schistocerca gregaria) is a destructive migratory pest, posing great threat to over 60 countries globally. In the backdrop of climate change, habitat suitability locusts poised undergo alterations. Hence, investigating shifting dynamics habitats holds profound significance in ensuring global agricultural resilience and food security. this study, we combined maximum entropy modelling geographic information system technology conduct comprehensive analysis impact change on...
Designing protein sequences that fold to a given three-dimensional (3D) structure has long been challenging problem in computational structural biology with significant theoretical and practical implications. In this study, we first formulated as predicting the residue type 3D environment around C α atom of residue, which is repeated for each protein. We designed nine-layer deep convolutional neural network (CNN) takes input gridded box atomic coordinates types residue. Several CNN layers...
Many proteins bear multi-locational characteristics, and this phenomenon is closely related to biological function. However, most of the existing methods can only deal with single-location proteins. Therefore, an automatic reliable ensemble classifier for protein subcellular multi-localization needed. We propose a new combining KNN (K-nearest neighbour) SVM (support vector machine) algorithms predict localization eukaryotic, Gram-negative bacterial viral based on general form Chou's pseudo...
With the rapid increase of protein sequences in post-genomic age, it is challenging to develop accurate and automated methods for reliably quickly predicting their subcellular localizations. Till now, many efforts have been tried, but most which used only a single algorithm. In this paper, we proposed an ensemble classifier KNN (k-nearest neighbor) SVM (support vector machine) algorithms predict localization eukaryotic proteins based on voting system. The overall prediction accuracies by...
Protein domain classification is an important step in metagenomic annotation. The state-of-the-art method for protein profile HMM-based alignment. However, the relatively high rates of insertions and deletions homopolymer regions pyrosequencing reads create frameshifts, causing conventional HMM alignment tools to generate alignments with marginal scores. This makes error-containing gene fragments unclassifiable tools. Thus, there a need accurate tool that can detect correct sequencing...
Gene assembly, which recovers gene segments from short reads, is an important step in functional analysis of next-generation sequencing data. Lacking quality reference genomes, de novo assembly commonly used for RNA-Seq data non-model organisms and metagenomic However, heterogeneous sequence coverage caused by expression or species abundance, similarity between isoforms homologous genes, large size all pose challenges to assembly. As a result, existing tools tend output fragmented contigs...
A (GGGGCC) hexanucleotide repeat (HR) expansion in the C9ORF72 gene, and its associated antisense (CCCCGG) expansion, are considered major cause behind frontotemporal dementia amyotrophic lateral sclerosis. We have performed molecular dynamics simulations to characterize conformation of 12 duplexes that result from three different reading frames sense HRs for both DNA RNA. These display atypical structures relevant not only a level understanding these diseases but also enlarging repertoire...
Sequences rich in glutamine (Q) and asparagine (N) are intrinsically disordered monomeric form, but can aggregate into highly ordered amyloids, as seen Q/N-rich prion domains (PrDs). Amyloids fibrillar protein aggregates β-sheet structures that self-propagate through protein-conformational chain reactions. Here, we present a comprehensive theoretical study of N/Q-rich peptides, including sequences found the yeast Sup35 PrD, parallel antiparallel aggregates, probe via fully atomistic...
Abstract Protein ligand docking is an indispensable tool for computational prediction of protein functions and screening drug candidates. Despite significant progress over the past two decades, it still a challenging problem, characterized by limited understanding energetics between proteins ligands, vast conformational space that has to be searched find satisfactory solution. In this project, we developed novel reinforcement learning (RL) approach, asynchronous advantage actor-critic model...
A (GGGGCC) hexanucleotide repeat (HR) expansion in the C9ORF72 gene has been considered major cause behind both frontotemporal dementia and amyotrophic lateral sclerosis, while a (GGGCCT) is associated with spinocerebellar ataxia 36. Recent experiments involving NMR, CD, optical melting 1D 1H NMR spectroscopy, suggest that r(GGGGCC) HR can adopt hairpin structure G-G mismatches equilibrium G-quadruplex structure. G-Quadruplexes have also identified for d(GGGGCC). As these lack molecular...
Sepsis is a leading cause of death over the world and septic shock, most severe complication sepsis, reaches mortality rate as high 50%. Early diagnosis treatment can prevent morbidity mortality. Nowadays, increasing availability electronic health records (EHRs) has generated great interests in developing models to predict acute medical conditions such shock. However, shock prediction faces two major challenges : 1) how capture informative progression long visit hospital patient; 2) obtain...
Abstract Predicting structures accurately for natural protein sequences by DeepMind’s AlphaFold is certainly one of the greatest breakthroughs in biology twenty-first century. For designed or engineered sequences, which can be unstable, predicting stabilities together with their essential since unstable will not function properly. We found that experimentally measured stability changes point mutations correlate poorly confidence scores produced AlphaFold. However, predicted using features...