手法を比較

	機械学習支援RNAシーケンス差次的発現解析	経路濃縮解析
分野	バイオインフォマティクス	バイオインフォマティクス
系統	Process / pipeline	Process / pipeline
提唱年	2015–2019 (rapid development period)	2003–2005
提唱者	Multiple groups; scVI (Lopez et al., 2018) and DCA (Eraslan et al., 2019) are landmark tools	Mootha et al. (2003); systematised by Subramanian et al. (2005)
種類	Computational bioinformatics pipeline	Statistical functional annotation method
原典	Lopez, R., Regier, J., Cole, M. B., Jordan, M. I., & Yosef, N. (2018). Deep generative modeling for single-cell transcriptomics. Nature Methods, 15(12), 1053–1058.	Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S., & Mesirov, J. P. (2005). Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences, 102(43), 15545–15550.
別名	ML-based DE analysis, deep learning RNA-seq DE, neural network differential expression, ML-augmented transcriptomics	PEA, overrepresentation analysis, ORA, functional enrichment analysis
関連	5	6
概要	Machine learning-assisted RNA-seq differential expression analysis augments classical statistical DE testing (DESeq2, edgeR, limma-voom) with ML models — including neural networks, random forests, and variational autoencoders — to better handle the high dimensionality, zero-inflation, and batch effects inherent in RNA-seq count data. The approach improves feature selection, noise reduction, and detection power, especially in large or complex experimental designs.	Pathway enrichment analysis (PEA) is a statistical approach that takes a list of genes or proteins of interest — typically derived from a differential expression or proteomics experiment — and identifies which pre-defined biological pathways or functional gene sets are represented more often than expected by chance. By mapping individual molecular changes onto curated pathway knowledge bases such as KEGG, Gene Ontology, or Reactome, PEA translates long gene lists into interpretable biological processes, making it a central tool in the post-analysis of high-throughput omics experiments.