Principal Components Analysis Based Unsupervised Feature Extraction Applied to Gene Expression Analysis of Blood from Dengue Haemorrhagic Fever Patients

Abstract

Dengue haemorrhagic fever (DHF) sometimes occurs after recovery from the disease caused by Dengue virus (DENV), and is often fatal. However, the mechanism of DHF has not been determined, possibly because no suitable methodologies are available to analyse this disease. Therefore, more innovative methods are required to analyse the gene expression profiles of DENV-infected patients. Principal components analysis (PCA)-based unsupervised feature extraction (FE) was applied to the gene expression profiles of DENV-infected patients, and an integrated analysis of two independent data sets identified 46 genes as critical for DHF progression. PCA using only these 46 genes rendered the two data sets highly consistent. The application of PCA to the 46 genes of an independent third data set successfully predicted the progression of DHF. A fourth in vitro data set confirmed the identification of the 46 genes. These 46 genes included interferon- and heme-biosynthesis-related genes. The former are enriched in binding sites for STAT1, STAT2, and IRF1, which are associated with DHF-promoting antibody-dependent enhancement, whereas the latter are considered to be related to the dysfunction of spliceosomes, which may mediate haemorrhage. These results are outcomes that other type of bioinformatic analysis could hardly achieve.

An unsupervised feature extraction and selection strategy for identifying epithelial-mesenchymal transition state metrics in breast cancer and melanoma

“…We also note that state metrics developed using microarray technology (e.g., (16)) are not likely relevant for interpreting data based on RNA se-quencing, given the unclear relation between transcriptome and protein abundance as assayed using microarray technology. While rarely used, the data-driven nature of unsupervised methods for feature extraction and selection are attractive (12). For instance, Umeyama et al used an unsupervised approach for feature extraction to identify genes associated with metastasis (31).…”

“…While effective, supervised methods can perform poorly if the strategy is based on misinformation, such as sample misclassification or prior biases as to the number of cell states or defining genes. While used less frequently, unsupervised methods for feature extraction and selection are advantageous as they can be data-driven (12). Here, our objective was to develop an unsupervised gene signature capturing this change in phenotype that is tailored to the specific cellular context of breast cancer and melanoma, as a illustrative example.…”

Drug candidate identification based on gene expression of treated cells using tensor decomposition-based unsupervised feature extraction for large-scale data

“…In this case, the 'PPI Hub Proteins' category in Enrichr was used. Labels (1) to (13) represent the same cell lines as described in Table 2. The full list of confusion matrices and commonly selected genes is available in Additional file 3…”

Tensor decomposition-Based Unsupervised Feature Extraction Applied to Single-Cell Gene Expression Analysis

“…The various unsupervised gene selection methods were invented, e.g., highly variable genes, bimodal genes and dpFeature and principal component analysis (PCA) based unsupervised feature extraction (FE) (50, 34, 49, 42, 25, 31, 27, 40, 47, 4, 5, 44, 45, 13, 12, 11, 51, 39, 43, 22, 23, 24, 26, 41, 48). Chen et al (2) recently compared genes selected by these methods and concluded that the genes selected are very diverse and have their own (unique) biological features.…”