Supervised Machine Learning Models and Protein-Protein Interaction Network Analysis of Gene Expression Profiles Induced by Omega-3 Polyunsaturated Fatty Acids

Background: Omega-3 polyunsaturated fatty acids (PUFAs), such as eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids, have beneficial effects on human health, but their effect gene expression in elderly individuals (age ≥ 65) is largely unknown. In order to examine this, the profiles were analyzed healthy subjects (n = 96) at baseline after 26 weeks of supplementation with EPA+DHA determine up-regulated down-regulated dif-ferentially expressed genes (DEGs) triggered by PUFAs. The protein-protein interaction (PPI) networks constructed mapping these DEGs a interactome linking them specific pathways. Objective: This study aimed implement supervised machine learning models network analysis induced Methods: transcriptional profile GSE12375 was obtained from Gene Expression Om-nibus database, which based Affymetrix NuGO array. probe cell intensity data converted into values, background correction performed multi-array average algorithm. LIMMA (Linear Models for Microarray Data) algo-rithm implemented identify relevant supplemen-tation p-value < 0.05. DAVID web server used construct en-riched KEGG (Kyoto Encyclopedia Genes Genomes) Finally, construction (ML) models, including logistic regression, naïve Bayes, deep neural networks, associated Results: results revealed that neurotrophin/MAPK signaling, whereas linked cancer, acute myeloid leukemia, long-term depression Additionally, ML approaches able cluster EPA/DHA-treated control groups regression performing best. Conclusion: Overall, this highlights pivotal changes PUFAs provides rationale implementation algorithms predictive type biomedical data.