Article
Biochemical Research Methods
He-Ming Chu, Xiang-Zhen Kong, Jin-Xing Liu, Chun-Hou Zheng, Han Zhang
Summary: This paper proposes a new preprocessing method and a new biclustering algorithm for processing gene expression data. Experimental results demonstrate the good performance of these two methods on synthetic and real datasets.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Review
Biochemical Research Methods
Kath Nicholls, Chris Wallace
Summary: This study compared four classes of sparse biclustering algorithms on simulated and real datasets, finding that Bayesian algorithms had high accuracy but were slower, while non-negative matrix factorisation algorithms performed poorly but could be repurposed for biclustering with post-processing. The study also highlighted the need to avoid differences dominating analyses in multi-tissue studies.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Biochemical Research Methods
He-Ming Chu, Jin-Xing Liu, Ke Zhang, Chun-Hou Zheng, Juan Wang, Xiang-Zhen Kong
Summary: This paper discusses biclustering algorithms for processing gene expression datasets, focusing on the handling of binary data. A new biclustering algorithm called AMBB is proposed, which constructs clusters based on an adjacency difference matrix. Experimental results demonstrate the high practicability of the algorithm when handling real datasets.
BMC BIOINFORMATICS
(2022)
Article
Health Care Sciences & Services
Weijie Zhang, Christine Wendt, Russel Bowler, Craig P. Hersh, Sandra E. Safo
Summary: In biomedical research, the integrative sparse singular value decomposition (iSSVD) method is effective in detecting meaningful biclusters in both single-view and multi-view data analyses.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2022)
Article
Engineering, Biomedical
Jianjun Sun, Qinghua Huang
Summary: This study proposes a multi-objective evolutionary algorithm with three populations to solve the biclustering problem, and introduces a novel bicluster seed generation method for better initialization. The experimental results demonstrate that the proposed method achieves better results under different noise levels and bicluster sizes.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Raghu Yelugam, Leonardo Enzo Brito da Silva, Donald C. Wunsch II
Summary: This study combines the adaptive resonance theory (ART)-based methods of biclustering ARTMAP (BARTMAP) and topological ART (TopoART) to produce TopoBARTMAP, which improves the quality of biclustering and module extraction. The capabilities of TopoBARTMAP were benchmarked using 35 real world cancer datasets and showed a statistically significant improvement over other assessed methods. The method also performed better at identifying different types of biclusters and represented gene bicluster associations.
Article
Biochemical Research Methods
Koyel Mandal, Rosy Sarmah, Dhruba Kumar Bhattacharyya
Summary: The paper introduces the POPBic algorithm, incorporating KEGG pathways to discover genes with similar expression patterns based on pathway relationships. Experimental results demonstrate the algorithm's sensitivity and robustness in the presence of noise, confirming its ability to detect biologically significant biclusters.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2021)
Article
Genetics & Heredity
Srinivas Rajagopalan, Amartya Singh, Hossein Khiabanian
Summary: The study utilized an unsupervised biclustering approach to analyze a large number of normal and tumor brain samples, identifying co-regulated gene expression profiles linked to specific subsets of gliomas and presenting a cilium-associated signature as predictor of poor survival in tumors. The introduction of a risk score based on expression of 12 cilium-associated genes proved to be informative of survival independently of other prognostic biomarkers, highlighting the potential therapeutic vulnerabilities for aggressive gliomas.
FRONTIERS IN GENETICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Koyel Mandal, Rosy Sarnnah, Dhruba Kumar Bhattacharyya, Jugal Kumar Kalita, Bhogeswar Borah
Summary: Effective biomarkers are crucial for early diagnosis and monitoring of breast cancer. Analyzing breast cancer miRNA data helps identify biomarkers that contribute to better understanding of the disease and making important clinical decisions. Our proposed rank-preserving biclustering algorithm outperforms other algorithms in relevance and recovery, showcasing its effectiveness in identifying biclusters for breast cancer data.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2021)
Article
Biochemical Research Methods
Qiong Fang, Dewei Su, Wilfred Ng, Jianlin Feng
Summary: The emergence of single-cell RNA sequencing techniques has opened up new opportunities for studying cell-specific changes in transcriptomic data. The challenge lies in effectively identifying cell subpopulations based on high-dimensional noisy scRNA-seq data. DivBiclust, a biclustering-based framework, demonstrates superior performance in accurately identifying cell subpopulations compared to existing methods.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2021)
Article
Genetics & Heredity
Andre Patricio, Rafael S. Costa, Rui Henriques
Summary: In this study, machine learning algorithms and high-throughput technologies were used to predict the treatment response of Hodgkin's lymphoma patients to multiagent chemotherapy. The proposed methodology demonstrated improved performance in predicting treatment response, providing valuable insights for improving patient outcomes.
BMC MEDICAL GENOMICS
(2023)
Article
Computer Science, Theory & Methods
R. Gowri, R. Rathipriya
Summary: The case study of plant intelligence inspired the development of novel NSI algorithms, namely Venus Flytrap Optimization and Bladder-Worts Suction, which are designed based on the prey-hunting mechanisms of Venus Flytrap and BladderWorts plants. These algorithms are compared and shown to be effective in extracting highly correlated maximal local patterns in gene expression data through biclustering. The NSI algorithms outperform existing optimization techniques like PSO and GA, as well as biclustering approaches like Cheng and Church, OPSM, BiMax, and Plaid approaches.
Article
Mathematical & Computational Biology
Yiran Zhang, Kellie J. Archer
Summary: In high-dimensional gene expression data, modeling for ordinal response helps to identify important genes for developing new diagnostic and prognostic tools for predicting or classifying stages of disease. A new Bayesian approach proposed in the study outperforms existing frequentist methods in simulation studies and is compared to frequentist methods in a study evaluating progression to hepatocellular carcinoma in hepatitis C infected patients.
STATISTICS IN MEDICINE
(2021)
Article
Biochemical Research Methods
Lulu Yan, Xiaoqiang Sun
Summary: This research comprehensively evaluates 14 deconvolution methods and proposes a new ensemble learning-based method, EnDecon, for more accurate deconvolution. The results show that cell2location, RCTD, and spatialDWLS are more accurate than other methods. Additionally, the study finds differences in the robustness of different methods to sequencing depth, spot size, and normalization choices, and most methods perform best when using the normalization method described in their original papers. This study provides valuable information and guidelines for the practical application of ST deconvolution tools and the development of new and more effective methods.
Article
Computer Science, Information Systems
Bhawani Sankar Biswal, Anjali Mohapatra, Swati Vipsita
Summary: In this paper, a novel ensemble biclustering approach called Ensemble Neighborhood search (ENS) is proposed based on the concept of neighborhood search. The simulation results demonstrate that the proposed approach is more flexible and adaptive compared to existing competitive methods on high-dimensional gene expression microarray data and scRNA-seq datasets. The proposed framework shows its effectiveness and computational efficiency in analyzing gene expression microarray data and high sparsity scRNA-seq data. The ENS algorithm is considered as a reliable framework and it can be used to improve the quality of biclusters.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)