Machine learning-based proteomics tool for assessing antibiotic mechanism of action
The perpetual development of antibiotic resistance necessitates the continuous discovery of novel antibiotic treatment options. Surprisingly, there are limited tools available to gain insights regarding an antibiotic’s mechanism of action (MoA). Currently, a critical knowledge gap exists regarding E. coli proteomic changes in response to antibiotics. We hypothesize that bacteria treated with antibiotics will have altered protein levels that are related to MoA. We propose developing a proteomics workflow, utilizing Escherichia coli (E. coli) as a model organism. Proteomes of antibiotic treated E. coli will be assessed using unsupervised clustering and supervised classification algorithms to correlate antibiotic MoA with proteomic response. As such, this project will fill a knowledge gap and establish a MoA platform through a hypothesis driven set of aims. Further development may result in a powerful tool to aid in identification of novel antibiotic MoA. Current progress is focused on protocol optimization.