Recovery of an 18 th Century Rhinovirus Genome through Ancient RNA Isolation of Human Lungs

Figures

• 

  Fig. 1.. Co-isolation of aDNA/aRNA from non-formalin fixed human lung samples. (A) Lung specimens GLAHM:120916 (left) and GLAHM:120917 (right) preserved in alcohol solution. (B) Concentrations of co-isolated RNA and DNA. Measurements reported from the Qubit high sensitivity RNA assay (black) and the high sensitivity dsDNA assay (gray). 100 uL of preservation liquid was taken from each specimen and used as an environmental control. 100 uL of water was processed as a negative control. (C) Histograms depicting length distributions of adapter trimmed RNA and DNA libraries. RNA (black) and DNA (gray) libraries of GLAHM:120916 (top row) and GLAHM:120917 (bottom row). (D) Distribution of human mapped reads across biotypes. Assignments relative to the hg38 GENCODE with repetitive elements (Supplementary Materials). Assignments are labeled in color. (E) Misincorporation frequencies of RNA and DNA reads relative to hg38. The plot displays the C>T (red), and G>A (blue) misincorporation frequencies (y-axis) calculated using mapDamage(29). Misincorporation rates of the complement strand are indicated by negative values on the x-axis.

• 

  Fig. 2.. Reconstruction of an 18th century human rhinovirus A genome. (A) Number of viral k-mers identified by Kraken2 on each lung specimen and control RNA libraries. K-mers per viral species were extracted using the Pavian suite, and all pathogens identified within the negative control library were filtered out of the subsequent libraries to remove spurious assignments. Viral species assignments are indicated by different colors and the total number of k-mers assigned per library are shown at the top of each bar. (B) Strand specific coverage of RNA reads mapping to the de novo assembled rhinovirus A genome. Positive strand coverage of the virus is displayed in blue, negative strand coverage is displayed in orange, and total coverage is represented in gray. Yellow vertical lines show gaps within the reads mapping to the assembled genome. MegaHit assembled contigs are represented by the gray bars at the top of the plot. A diagram of the HRV genome is displayed at the bottom of the plot. (C) Damage pattern of GLAHM:120916 RNA reads mapping to the de novo assembled HRV A genome. The plot displays the C>T (red), and G>A (blue) misincorporation frequencies (y-axis) calculated using…

• 

  Fig. 3.. Evolutionary characterization of the 18th century HRV A genome: (A) Amino acid (AA) variation in the historical HRV-A genome. The bottom panel shows AA differences across the CDS between the historical HRV A and any HRV A19 (green lines) and AA positions that are unique to the historical HRV A relative to all HRV A sequences (purple lines). Green line height indicates the proportion of HRV A19 sequences differing at each position. Historical HRV A genome coverage is shown in the background, with coding sequence annotations below. The top panel shows Shannon entropy per AA position (blue). Purple arrows mark the two historical-specific AAs with near-zero entropy (~0.01). Dashed and dotted red lines denote highly (>2) and moderately (>1) variable positions, respectively. (B) BETS analyses displaying the log Bayes Factor (ΔMLE) from BEAST trees including tip dates (heterochronous) and those not including tip dates (isochronous). The Marginal Likelihood Estimates are calculated from BEAST trees utilizing a Generalized Stepping-Stone Sampling analyses. A ΔMLE greater than 3 is sufficient to assume strong temporal signal(44). The 95% HPD of the mean substitution rate and root age are…