Triple-negative breast cancer (TNBC), characterized by swift progression, high metastasis and recurrence rates, and limited treatment options, was analyzed to determine specific genes associated with metabolic reprogramming to form unique biomarkers. Utilizing machine learning algorithms, the researchers screened 2 TNBC datasets (GEO and TCGA) to identify differentially expressed genes linked to metabolic reprogramming prior to extracting key genes in a variety of cohorts. Transcription factors and ceRNA (circular RNA) targeting these key genes were then found before comparing to immunohistochemical staining. The researchers found distinct expression and regulation patterns correlating to specific signaling and maintenance genes, many tied to receptors and protein interactions. The machine learning model was employed to assess these expression levels and their relationship to clinical values. Certain genes were found to correlate with specific immune responses with long-term inactivation of memory CD4+ T cells being a primary cause for metastasis. In conclusion, this machine-learning algorithm is crucial to determining specific receptors and mechanisms associated with TNBC progression as confirmed by clinical symptoms and immunohistochemical staining.