Person:

Lemieux, Jacob

Parasitic protozoan infections of the red blood cell are among the most widespread and devastating pathogens of vertebrates. In humans, two genera of pathogens cause disease: the Plasmodia, which cause malaria, and the Babesia, which cause babesiosis. In this thesis, we apply the tools of whole genome sequencing and evolutionary genetics to study factors contributing to the spread of these pathogens: in P. falciparum, the acquisition of multiple drug resistance, and in B. microti the development of azithromycin resistance and the population genetics of emergence. In the first part of this work, we test whether an accelerated mutation rate predisposes to acquisition of drug resistance in P. falciparum. Epidemiologically, resistance tends to begin along the Thai-Cambodian border, and from there spreads to other parts of the world. Environmental conditions such as inadequate drug dosing likely facilitate drug resistance, but molecular evidence also suggests that parasites from the Thai-Cambodian border may harbor genetic traits that let them develop resistance to novel antimalarials at an elevated rate. Low-dose drug pressure has also been proposed to be mutagenic, since several antimalarial agents have known DNA binding properties and have been shown to impair DNA damage repair pathways in P. falciparum. To test these hypotheses, we directly assayed substitution rates in a parasite line from the Thai-Cambodian border and a South American isolate, with and without chloroquine pressure. Sampling parasite DNA over a total of 760 generations (~4.2 years), we identified 17 mutations, producing an estimate of the substitution rate at 1.065x10^9 substitutions per site, per generation. We find that chloroquine pressure does not alter the mutation rate. We further find that substitutions accrued at an approximately 3-fold rate in the lines from Southeast Asia, a result which trended toward but did not reach statistical significance (p = 0.056). We argue that this is insucient by itself to account for the rapidly increased rate at which ARMD parasites acquire drug resistance. By sequencing intermediate timepoints, we also characterize the dynamics of allele substitution in vitro. In the second part of this thesis, we characterize Babesia microti by sequencing clinical isolates and enzootic strains. Since the first case in 1969 [36], human babesiosis due B. microti has emerged as important infection in the Northeast USA [84]. In order to characterize natural selection, recent evolutionary history, and the genetic architec ture of Babesia microti populations, we created a map of genetic diversity from clinical strains. We describe this map, and show that B. microti isolates from the Northeast USA possess a paucity of nucleotide diversity, consistent with very recent common ancestry of circulating strains. We describe how B. microti genomes display a predominance of rare alleles and a number of segregating sites in excess of pairwise nucleotide diversity, suggestive of a recent population expansion. Finally, we identify RPL4 as a candidate gene for azithromycin resistance based on a non-synonymous substitution that occurs in a highly conserved arginine in the azithromycin binding region of the L4 component of the 50S ribosomal subunit in a patient with azithromycin failure.

Abstract We describe a patient with severe and progressive encephalitis of unknown etiology. We performed rapid metagenomic sequencing from cerebrospinal fluid and identified Powassan virus, an emerging tick-borne flavivirus that has been increasingly detected in the United States.

The global spread and continued evolution of SARS-CoV-2 has driven an unprecedented surge in viral genomic surveillance. Amplicon-based sequencing methods provide a sensitive, low-cost and rapid approach but suffer a high potential for contamination, which can undermine laboratory processes and results. This challenge will only increase with expanding global production of sequences by diverse laboratories for epidemiological and clinical interpretation, as well in genomic surveillance in future outbreaks. We present SDSI+AmpSeq, an approach which uses synthetic DNA spike-ins (SDSIs) to track samples and detect inter-sample contamination through the sequencing workflow. Applying SDSIs to the ARTIC Consortium’s amplicon design, we demonstrate their utility and efficiency in a real-time investigation of a suspected hospital cluster of SARS-CoV-2 cases and across thousands of diagnostic samples at multiple laboratories. We establish that SDSI+AmpSeq provides increased confidence in genomic data by detecting and in some cases correcting for relatively common, yet previously unobserved modes of error without impacting genome recovery.