A5101612790- Machine Learning in Bioinformatics
- Fractal and DNA sequence analysis
- Gene expression and cancer classification
- RNA and protein synthesis mechanisms
- Bioinformatics and Genomic Networks
- Data Mining Algorithms and Applications
- Genomics and Phylogenetic Studies
- Gene Regulatory Network Analysis
- AI in cancer detection
- Mitochondrial Function and Pathology
- Machine Learning and ELM
- Scientific Research and Philosophical Inquiry
- Library Science and Information
- Information Systems and Technology Applications
- Neuroscience and Neuropharmacology Research
- Text and Document Classification Technologies
- Electron Spin Resonance Studies
- Bone Metabolism and Diseases
- Rough Sets and Fuzzy Logic
- Physical Unclonable Functions (PUFs) and Hardware Security
- Advanced X-ray and CT Imaging
- Neuroscience and Neural Engineering
- Imbalanced Data Classification Techniques
- Face and Expression Recognition
- Advanced Multi-Objective Optimization Algorithms
Hunan University
2006-2018
Air Force Medical University
2015
Abstract This study evaluated the association between free fatty acid (FFA), ROS generation, mitochondrial dysfunction and bone mineral density (BMD) in type 2 diabetic patients investigated molecular mechanism. db/db high fat (HF)-fed mice were treated by Etomoxir, an inhibitor of CPT1, MitoQ PFT-α, P53. Bone metabolic factors assessed BMSCs isolated induced to osteogenic differentiation. FFA, lipid peroxidation mtDNA copy number correlated with BMD T2DM patients. PFT-α significantly...
Abstract Based on chemical properties of the neighboring dual nucleotides, we reduce a DNA sequence into four 3D graphical representations. Associating with eigenvalues introduced covariance matrix and measure similarity, introduce an approach to make similarity analysis sequence. © 2008 Wiley Periodicals, Inc. Int J Quantum Chem,
Abstract On the basis of information on evolution 20 amino acids and their physiochemical characteristics, we propose a new two‐dimensional (2D) graphical representation protein sequences in this article. By method, use 2D data to represent three‐dimensional constructed by acids' index, class acid based order appearing sequences. Then, using discrete Fourier transform, sequence signals with different lengths can be transformed frequency domain, which are same length. A method is used analyze...
Abstract A two‐dimensional graphical representation (2DGRR) of RNA secondary structures using a two Cartesian coordinates system has been derived for mathematical denotation structure. The 2DGRR resolves structure degeneracy and avoids loss information the limitation that different correspond to same curve. pseudo‐knots also can be represented as 2D representations. © 2006 Wiley Periodicals, Inc. Int J Quantum Chem,
Inferring gene regulatory networks from expression data is a central problem in systems biology.
In the present study, we introduce a novel semi-supervised method called maximum discriminative local margin (semiMM) for gene selection in expression data. The semiMM is "filter" approach that exploits structure, variance, and mutual information. We first constructed nearest neighbour graph divided this information into within-class between-class graphs by weighing edge between two data points. aims to discover most features classification via maximizing data, variance of all with class...
According to the three classifications of nucleotides, we introduce a sort binary coding method RNA secondary structures. On basis this representation, can reduce structure into digit sequences. We also propose rules based on exclusive-OR operation. Associating with proposed rules, judge mutation between bases or base and pair, make sequence alignment easily.
Abstract The classification of tumors is crucial for the proper treatment cancer. Sparse representation-based classifier (SRC) exhibits good performance and has been successfully used to classify using gene expression profile data. In this study, we propose a three-step maxdenominator reweighted sparse representation (MRSRC) method tumors. First, extract set metagenes from training samples. These can capture structures inherent data are more effective than original Second, use "Equation...
Time-course gene expression datasets, which record continuous biological processes of genes, have recently been used to predict function. However, only few positive genes can be obtained from annotation databases, such as ontology (GO). To obtain more useful information and effectively function, annotations are clustered together form a learnable effective learning system. In this paper, we propose novel multi-instance hierarchical clustering (MIHC) method establish system by GO compare with...
Tumor classification is crucial to the clinical diagnosis and proper treatment of cancers. In recent years, sparse representation-based classifier (SRC) has been proposed for tumor classification. The employed dictionary plays an important role in or coding-based However, models have not used dictionary, thereby limiting their performance. Furthermore, this representation model assumes that coding residual follows a Gaussian Laplacian distribution, which may effectively describe practical...
Gene function annotation is the main challenge in post genome era, which an important part of annotation. The sequencing human project produces a whole data, providing abundant biological information for study gene However, to obtain useful knowledge from large amount potential strategy apply machine learning methods mine these data and predict function. In this study, we improved multi-instance hierarchical clustering by using ontology hierarchy annotate function, combines with multi-label...