A5070302649- Gene expression and cancer classification
- Bioinformatics and Genomic Networks
- Machine Learning in Bioinformatics
- Computational Drug Discovery Methods
- Face and Expression Recognition
- Neural Networks and Applications
- Evolutionary Algorithms and Applications
- Gene Regulatory Network Analysis
- Genetics, Bioinformatics, and Biomedical Research
- Rough Sets and Fuzzy Logic
- Genomics and Phylogenetic Studies
- Algorithms and Data Compression
- Machine Learning and Data Classification
- AI in cancer detection
- Bayesian Modeling and Causal Inference
- Wireless Communication Networks Research
- Blind Source Separation Techniques
- Protein Structure and Dynamics
- Sparse and Compressive Sensing Techniques
- Optimal Experimental Design Methods
- Genetic Mapping and Diversity in Plants and Animals
- Cell Image Analysis Techniques
- Biomedical Text Mining and Ontologies
- Tensor decomposition and applications
- Advanced Wireless Network Optimization
University of Windsor
2014-2024
Ho Chi Minh City University of Industry and Trade
2019
Harbin Institute of Technology
2019
Dalat University
2016-2019
Australian National University
2010
University of Ottawa
1998-2005
Université du Québec à Trois-Rivières
2005
Lakehead University
2000-2005
Non-negative matrix factorization (NMF) has been introduced as an important method for mining biological data. Though there currently exists packages implemented in R and other programming languages, they either provide only a few optimization algorithms or focus on specific application field. There does not exist complete NMF package the bioinformatics community, order to perform various data tasks data.We convenient MATLAB toolbox containing both implementations of techniques variety...
Studying breast cancer survivability among different patients who received various treatment therapies may help us understand the relationship between and of based on genetic expression. In this work, we present a classification system that predicts whether given patient underwent through hormone therapy, radiotherapy, or surgery will survive beyond five years after treatment. Our classifier is tree-based hierarchical approach groups classes. Each node in tree associated with therapy subset...
Large Language Models (LLMs) like Generative Pre-trained Transformer (GPT) from OpenAI and LLaMA (Large Model Meta AI) AI are increasingly recognized for their potential in the field of cheminformatics, particularly understanding Simplified Molecular Input Line Entry System (SMILES), a standard method representing chemical structures. These LLMs also have ability to decode SMILES strings into vector representations.
In the context of highly scalable distributed resource management architectures for grid computing, we present a genetic algorithm based scheduler. A scheduler must use available resources efficiently, while satisfying competing and mutually conflicting goals. The workload may consist multiple jobs, with quality-of-service constraints. directed acyclic graph (DAG) represents each job, taking into account arbitrary precedence constraints processing time. has been designed to be compatible...
High-throughput genomic and proteomic data have important applications in medicine including prevention, diagnosis, treatment, prognosis of diseases, molecular biology, for example pathway identification. Many such can be formulated to classification dimension reduction problems machine learning. There are computationally challenging issues with regards accurately classifying data, which due dimensionality, noise redundancy, name a few. The principle sparse representation has been applied...
Drug repurposing is a potential alternative to the traditional drug discovery process. can be formulated as recommender system that recommends novel indications for available drugs based on known drug-disease associations. This article presents method non-negative matrix factorization (NMF-DR) predict drug-related candidate disease indications. work proposes system-based by integrating and diseases related data sources. For this purpose, framework first integrates two types of similarities,...
In wireless mobile communication systems, radio spectrum is a limited resource. However, efficient use of available channels has been shown to improve the system capacity. The role channel assignment scheme allocate cells or mobiles in such way as minimize call blocking dropping probabilities, and also maximize quality service. Channel known be an NP-hard optimization problem. this paper, we have developed evolutionary strategy (ES) which optimizes assignment. proposed ES approach uses...
Microarray data can be used to detect diseases and predict responses therapies through classification models. However, the high dimensionality low sample size of such result in many computational problems as reduced prediction accuracy slow speed. In this paper, we propose a novel family nonnegative least-squares classifiers for high-dimensional microarray gene expression comparative genomic hybridization data. Our approaches are based on combining advantages using local learning,...
<ns4:p>Genomic aberrations and gene expression-defined subtypes in the large METABRIC patient cohort have been used to stratify predict survival. The present study normalized expression signatures of paclitaxel drug response outcome for different survival times patients receiving hormone (HT) and, some cases, chemotherapy (CT) agents. This machine learning method, which distinguishes sensitivity vs. resistance breast cancer cell lines validates predictions patients; was also derive other HT...
Non-negative information can benefit the analysis of microarray data. This paper investigates classification performance non-negative matrix factorization (NMF) over gene-sample We also extends it to higher-order version for clinical time-series data represented by tensor. Experiments show that NMF and achieve at least comparable prediction performance.
Accurately reconstructing gene regulatory network (GRN) from expression data is a challenging task in systems biology. Although some progresses have been made, the performance of GRN reconstruction still has much room for improvement. Because many events are asynchronous, learning interactions with multiple time delays an effective way to improve accuracy reconstruction. Here, we propose new approach, called Max-Min high-order dynamic Bayesian (MMHO-DBN) by extending hill-climbing technique...
Graph Neural Networks (GNNs) have emerged as powerful tools for analyzing structured data, particularly in domains where relationships and interactions between entities are key. By leveraging the inherent graph structure datasets, GNNs excel capturing complex dependencies patterns that traditional neural networks might miss. This advantage is especially pronounced field of computational biology, intricate connections biological play a crucial role. In this context, Our work explores...
Non-negative factorization (NMF) has been a popular machine learning method for analyzing microarray data. Kernel approaches can capture more non-linear discriminative features than linear ones. In this paper, we propose novel kernel NMF (KNMF) approach feature extraction and classification of Our is also generalized to high-order (HONMF). Extensive experiments on eight datasets show that our generally outperforms the traditional existing KNMFs. Preliminary experiment data shows KHONMF...
'De novo' drug discovery is costly, slow, and with high risk. Repurposing known drugs for treatment of other diseases offers a fast, low-cost/risk highly-efficient method toward development efficacious treatments. The emergence large-scale heterogeneous biomolecular networks, molecular, chemical bioactivity data, genomic phenotypic data pharmacological compounds enabling the new area repurposing called 'in silico' repurposing, i.e., computational (CDR). aim CDR to discover indications an...
Modern data generated in many fields are a strong need of integrative machine learning models order to better make use heterogeneous information decision making and knowledge discovery. How from multiple sources incorporated system is key step for successful analysis. In this paper, we provide comprehensive review on integration techniques perspective.
Abstract Background Mutations in the mitochondrial genome (mtgenome) have been associated with many disorders, including breast cancer. Nipple aspirate fluid (NAF) from symptomatic women could potentially serve as a minimally invasive sample for cancer screening by detecting somatic mutations this biofluid. This study is aimed at 1) demonstrating feasibility of NAF recovery women, 2) examining sequencing entire samples, 3) cross validation Human resequencing array 2.0 (MCv2), and 4)...
Purpose: Large Language Models (LLMs) like ChatGPT and LLaMA are increasingly recognized for their potential in the field of cheminformatics, particularly interpreting Simplified Molecular Input Line Entry System (SMILES), a standard method representing chemical structures. These LLMs can decode SMILES strings into vector representations, providing novel approach to understanding graphs. Methods: We investigate performance embedding strings. Our evaluation focuses on two key applications:...
<ns4:p>Genomic aberrations and gene expression-defined subtypes in the large METABRIC patient cohort have been used to stratify predict survival. The present study normalized expression signatures of paclitaxel drug response outcome for different survival times patients receiving hormone (HT) and, some cases, chemotherapy (CT) agents. This machine learning method, which distinguishes sensitivity vs. resistance breast cancer cell lines validates predictions patients, was also derive other HT...