10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Novel Method for Identification of Airborne Transmission Using Molecular Epidemiology
Donald Milton, Daniel J. Nasko, TODD TREANGEN, University of Maryland
Abstract Number: 1528 Working Group: Infectious Bioaerosol
Abstract We subjected three pairs of fine-aerosol and NP swab samples (from our archive of 188 paired samples with detectable viral RNA in aerosol samples) to deep sequencing as described below (except that for exploratory analysis we used 20 samples per MiSeq lane). This preliminary sequencing effort tested the feasibility of the proposed work by selecting aerosol samples covering the range of measured fine-aerosol viral RNA content from ~103 to 107 copies per 1 mL. We achieved an average read depth of 15,000X per sample (ranging from 12,000X to 20,000X). All variants with p-value < 0.05 and at allele frequency >=2% were considered true variants; the rest were discarded as putative false positives. This cutoff level is consistent with those used in several recent analyses of the transmission bottleneck for influenza and high enough to exclude variants that would be unlikely to be transmitted through a bottleneck of 100. For each of the six samples sequenced for the feasibility study (three pairs of NP swab and aerosol), we identified the set of single nucleotide polymorphisms (SNPs) unique to one sample compared across all samples, unique to each member of the pair within the pair. The number of variants found in one but not both members of each pair ranged from 20 to 147 with no shared variants; thus, Jaccard similarity coefficients were zero. Phylogenetic analysis demonstrated that two of the paired samples were very closely related with consensus sequences differing by fewer than 10 SNPs. The third subject’s consensus sequences from two sites differed by 73 SNPs, raising a question of two separate infection events at different sites. Overall, these preliminary data show that we can detect and distinguish variants and consensus sequences from the NP swab and the fine particle aerosol samples. If these results are confirmed by sequencing of additional samples, we will be able to establish that viral populations in the NP swabs and aerosol samples are independent. Success with these few samples suggests that we will be able to use the differences in consensus sequence and variant populations as markers of whether transmitted virus between two cases was from viral aerosol or a nasal source.