Current Research Directions

  • Viral dynamics of SARS-CoV-2
  • Microbial metagenomics using long-read  Nanopore sequencing
  • Viral metagenomics of upper respiratory RNA viruses
  • Carriage and disease dynamics of Staphylococcus aureus among Native Americans
  • Population dynamics of Streptococcus pneumoniae serotype 3 carriage and disease

Research Areas of Interest

Genomic Epidemiology and viral population dynamics of SARS-CoV-2.

Genomic Epidemiology of methicillin-resistant Staphylococcus aureus (MRSA).

MRSA remains an important pathogen in the healthcare setting as well as the community at large. Historically, available molecular typing techniques have limited the resolution at which S. aureus transmission could be investigated. The advent of whole-genome sequencing has facilitated the investigation of bacterial pathogens on a relatively short time frame (days to weeks) relevant for epidemiological investigations. In addition, genomic data may be used to study virulence, pathogenesis, population structure, and evolutionary history. My research has focused on the transmission dynamics and population genomics of MRSA in healthcare settings and at the community-hospital interface. Specifically, I have investigated the transmission of MRSA in neonatal intensive care units, finding that the majority of cases among hospitalized neonates were the result of multiple introductions of MRSA from the community, leading to hospital transmission chains. This finding had direct implications for infection control interventions. Most recently, I studied the intrahost evolution of MRSA among individuals with recurring skin and soft tissue infections. I found that antibiotic resistance emerged from both replacements of a susceptible population and acquisition of mobile genetic elements harboring resistance.


Population dynamics of pneumococcal disease.

Invasive disease caused by the bacterium Streptococcus pneumoniae, referred to as pneumococcal disease, remains a significant cause of morbidity and mortality worldwide, despite the availability of an effective vaccine. The persistence of pneumococcal disease stems in part from its population structure, which is comprised of several lineages that are strongly correlated with the bacterium’s capsular polysaccharide serotype. The vaccine targets only specific serotypes and as a result, after vaccine introduction, non-vaccine types fill the gap left by vaccine types. The processes by which this occurs and the evolutionary driving forces that generate and maintain pneumococcal population diversity are the focus of my postdoctoral research. To date, we have published the findings of our analysis that tested whether individuals’ antibodies to diverse pneumococcal protein antigens predict which strain they will subsequently be colonized with. This study was a collaboration between   Johns Hopkins University Center for American Indian Health, Wellcome Trust Sanger Institute, and University College London. Now, I am working on a larger study of pneumococcal population dynamics leveraging ~1000 pneumococcal genomes carried by Native Americans spanning a 12 year period. In addition, as a continuation of the collaboration with the Sanger Institute, we are analyzing ~350 international pneumococcal genomes from the Global Pneumococcal Sequencing project. Together, these studies are addressing important questions regarding protein antigen vaccine feasibility, the future of conjugate vaccine development, and the larger evolutionary dynamics of bacterial pathogens.


Population genomics of Vibrio cholerae in Hispaniola.

In 2010, Haiti was subject to a series of natural disasters followed by the introduction of epidemic Vibrio cholerae, which had not been documented in Hispaniola for over 100 years. Using genomic data from clinical and aquatic environmental isolates collected in Haiti from 2010-2013, I studied the evolution and diversification of cholera as it emerged in Haiti. I illustrated the environmental dissemination of toxigenic V. cholerae O1 suggesting ongoing transmission between the aquatic environments and susceptible hosts. Assessment of evolutionary selective pressures demonstrated a progressive increase of non-synonymous substitutions, suggesting diversification likely was driven by positive selection. This finding had implications for virulence, transmission dynamics, and even vaccine efficacy. Subsequently, during environmental sampling in Haiti, we identified two non-toxigenic O1 isolates that were phylogenetically related to the O1 classical strain. Further genomic analysis of these strains with other historical toxigenic V. cholerae genomes showed that they shared a most recent common ancestor approximately ~1500 AD, suggesting that cholera may have been introduced to Hispaniola at this time. In addition, they possessed the necessary genetic machinery to receive the CTX phage and become toxigenic strains. This is a concern as these strains intermix with the current epidemic strain in the aquatic environment.


Phylodynamics and phylogeography of fast-evolving viral pathogens.

Phylogenetic analysis is a proven method for investigating the population demographic history of fast involving viruses. I have been involved in two studies investigating the population dynamics of viral pathogens utilizing Bayesian phylogeography. In the first study, I assisted in understanding the distribution of HIV subtypes in Morocco using viral sequencing data in conjunction with national surveillance data. This study found that HIV subtype distribution was compartmentalized among epidemiological risk groups and varied geographically. In the second study, I investigated the spatial dispersion of Ebola Zaire specifically assessing selection for advantageous mutation from 1976 to the current epidemic. I found that based on historical genomic data, the emergence of an advantageous mutation that radically changes the mode of transmission was possible but unlikely. Overall, both of these studies had direct application for ongoing management of HIV and EBOV epidemics and demonstrated the power of phylogenetic and population dynamic methods.


%d bloggers like this: