Direct sequence repeats adjacent to DVG breakpoints. (A) Example of a DVG with a direct sequence repeat of 3 nt adjacent to the breaking point (indicated in blue). (B) Conditions for a jump from position a to position b to create a DVG with an uncertainty ≥1 (left diagram) or ≥2 (right diagram). (C,D) Distribution function (C) and complementary cumulative distribution function (D) of the uncertainty value for DVGs with random breakpoints, either computed from a random sequence (gray bars and line) or computed from the sequence of each segment (colored bars and line). (E–H) Distribution function (E,G) and complementary cumulative distribution function (F,H) of uncertainties of DVGs obtained for each indicated virus upon RT-seq (E,F) or RT-PCR-seq (G,H) analysis, compared with random DVGs computed for each sequence. (*) P < 0.05, (***) P < 0.001. Significance was determined using χ2 goodness-of-fit test, followed by multiple testing correction using Holm's method. (I) Distribution of each nucleotide within uncertainty sequences in DVGs with uncertainty = 1, ≤2, ≤3, or ≤4. Significance was determined using χ2 goodness-of-fit test, with the null distribution being computed from the A/WSN/33 full-length genome reference sequences, as shown in Figure 5C.
