Identifying Risk Factors for Antibiotic Resistance via Integration of Epidemiology and Metagenomics

PROJECT SUMMARYGiven the growing burden of antimicrobial resistance (AR) and lack of effective therapies for multi-drugresistant organisms, the development of new tools or models which risk-stratify patients for colonization andinfection by AR bacteria is of paramount importance, particularly in high-risk populations. The significance ofthe gut microbiome in mediating colonization resistance against drug resistant pathogens as well as the role ofmicrobiota-depleting antibiotics in the development of AR infections is being increasingly appreciated.However, there is currently a deficiency of methods integrating microbiome and antibiotic factors into AR-predictive algorithms. Thus, the overall objective of the proposed research is to improve understanding of thefactors driving the epidemiology of AR-colonization and infection by incorporating metagenomic and antibioticadministration data of a well-defined clinical cohort. In this proposal, we focus on patients with acutemyelogenous leukemia (from whom we have already collected extensive longitudinal stool samples andperformed 16S rRNA gene sequencing) because of the high rates of AR pathogen colonization and severe riskfor infection. The overarching hypothesis that will be tested is that the baseline presence of a limited numberof key bacterial species and antibiotic resistance genes (ARGs) are critical for the risk of colonization and/orinfection with an AR pathogen when combined with the administration of specific antimicrobials. We will beginour research by comprehensively determining the epidemiology of AR pathogen colonization and AR infectionin our cohort via culture based stool sample analyses and clinical chart review, respectively. Using shotgunmetagenomics, we will establish whether the baseline intestinal microbiome species and resistomecharacteristics are associated with the acquisition of AR pathogens colonizing or causing infection. Similarly,we will ascertain the relationship between antimicrobial exposure, microbiome disruption, and subsequent ARemergence. The data from these studies will be integrated into Decision Tree (DT) and Random Forest (RF)models to improve the prediction of AR pathogen colonization and AR infection outcomes. The proposedcareer development award, which utilizes the expertise of a superlative mentorship team and a uniquelydesigned research and training plan, will enable me the opportunity to build upon my current expertise inmicrobiology, genomics, and molecular epidemiology by adding advanced training in shotgun metagenomicanalyses, bioinformatics, and biostatistical modeling. Moreover, the numerous resources and support providedby my institution and mentoring team will ensure my successful transition to an independent investigator aswell as establish a strong foundation for my long-term goals of understanding and mitigating the impact ofantimicrobial resistance in human health via integration of multiple ?omics platforms and provision ofpersonalized genomic-based medicine.