Deciphering Stem Cell Pluripotency Using a Machine Learning Clustering Approach

Complex biological systems including genes, transcription factors, regulators, and signaling pathways control pluripotency. Based on their capacity for self-renewal, pluripotent stem cells can be divided into two categories: naive and primed. Rather than being static states, these stages depict a spectrum. Post-implantation embryonic stem cells that have undergone priming have little capacity for differentiation. This study uses gene expression data from unprimed and primed pluripotent embryonic stem cells to propose a novel clustering approach based on machine learning to evaluate stem cell pluripotency. Unique clusters of genes with varied degrees of pluripotency are found and verified by examining publicly accessible gene expression data and carrying out functional enrichment studies. Gene ontology (GO) analysis, which finds enhanced biological processes, molecular functions, and cellular components associated with gene sets, necessitates specialized tools like clusterProfiler for gene expression analysis. It aids in comprehending how particular circumstances can influence certain processes. Another key component is network analysis, and Cytoscape aids in visualizing and analyzing intricate gene/protein networks.