Related Experiment Videos
Biomarker identification and cancer classification based on microarray data using Laplace naive Bayes model with mean
Meng-Yun Wu1, Dao-Qing Dai, Yu Shi
1Center for Computer Vision and Department of Mathematics, Sun Yat-Sen University,Guangzhou 510275, China. mc04wmy@mail2.sysu.edu.cn
Summary
This study introduces a robust Laplace naive Bayes model with mean shrinkage for accurate cancer classification and biomarker discovery. The method effectively handles gene expression data outliers and selects relevant genes, improving diagnostic capabilities.
Area of Science:
- Computational Biology
- Bioinformatics
- Machine Learning
Background:
- Biomarker identification and cancer classification are critical in genomics.
- Gene expression data often exhibits high gene correlation and outliers, complicating analysis.
- Existing methods may not adequately address group effects or outlier robustness.
Purpose of the Study:
- To develop a novel gene selection method for cancer classification.
- To enhance robustness against outliers and account for group effects in gene expression data.
- To identify reliable biomarkers for cancer diagnosis.
Main Methods:
- Proposed a Laplace naive Bayes model with mean shrinkage (LNB-MS).
- Utilized Laplace distribution for outlier resilience.
- Implemented L1 penalty for automatic feature selection and an efficient parameter estimation algorithm.
- Introduced a strategy to control the regularization parameter using the number of selected features.
Main Results:
- Demonstrated accuracy, sparsity, efficiency, and robustness on simulated and 17 cancer datasets.
- Identified numerous biomarkers subsequently verified in biomedical research.
- Showcased selection of highly correlated genes using Gene Ontology (GO) term analysis.
Conclusions:
- The LNB-MS model offers a robust and efficient approach for biomarker discovery and cancer classification.
- The method successfully addresses challenges of gene correlation and data outliers.
- Validated biomarkers and biologically relevant gene selection confirm the method's utility.
Related Concept Videos
01:21Cancer Survival Analysis
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
02:39lncRNA - Long Non-coding RNAs
In humans, more than 80% of the genome gets transcribed. However, only around 2% of the genome codes for proteins. The remaining part produces non-coding RNAs which includes ribosomal RNAs, transfer RNAs, telomerase RNAs, and regulatory RNAs, among other types. A large number of regulatory non-coding RNAs have been classified into two groups depending upon their length – small non-coding RNAs, such as microRNA, which are less than 200 nucleotides in length, and long non-coding RNA (lncRNA)...