Bacterial persisters are a multidrug-tolerant subpopulation capable of surviving and resuscitating after lethal antibiotic treatment, leading to relapsing infections and the emergence of antibiotic resistance. We hypothesize that multiple persister subpopulations are within an isogenicEscherichia colipopulation, allowing them to survive lethal antibiotic stress. We conducted transcriptomic analyses at multiple time points with lethal ampicillin (Amp) antibiotic, and as expected, several genes were differentially expressed over time. We identified a subset of genes consistently upregulated by comparing transcription levels at different time points of Amp-treated to untreated. Some genes had previously been associated with persisters, while others were new. Subsequently, network analysis showed the gene response between networks, but could not map hypothetical genes. So, we overexpressed seven hypothetical genes, which resulted in slow growth or no growth, indicating that high production harmed the cell. We then made single gene knockouts, which dramatically reduced persister level by ∼4-6 fold at 3 h and ∼10-15 fold at 6 h of Amp treatment. However, no significant difference in survival rates was observed at 24 h, indicating the presence of multiple persister subpopulations. Our mathematical model demonstrated a 20-fold decrease in the slow-decaying fraction in the mutant, suggesting the importance of decay kinetics in bacterial survival. These results support the existence of multiple subpopulations of persisters, each characterized by distinct decay rates. These results challenge the idea of complete dormancy, suggest the presence of intricate, multifaceted survival mechanisms, and indicate that the persister population itself is heterogeneous.Significance statementBacterial persisters, a subpopulation known for their multidrug tolerance and ability to survive lethal antibiotic treatments, have long posed challenges in understanding their formation and long-term survival. They are a driving force of antibiotic resistance, so it is paramount that we learn more about them as the antibiotic resistance problem continues to grow. Our study challenges the long-held consensus that persisters are completely dormant and are of one single population. Our results clearly show that persisters are not as dormant as once thought, and multiple populations of persisters form during lethal antibiotic treatment despite the cells being genetically identical. We use wet lab and dry lab (mathematical modeling) to demonstrate these new findings.