Altered sequencing success at non-B-DNA motifs

Modern sequencing technologies are not error-free, and might possess systematic biases in their error distributions. A potential cause for non-randomly occurring errors is the formation of alternative DNA structures (non-B DNA), such as G-quadruplexes (G4s), Z-DNA, or cruciform structures, during sequencing. Approximately 13% of the human genome has the potential to form such structures, which have been previously shown to affect the speed and accuracy of DNA polymerases. To test whether motifs with the potential to form non-B DNA (non-B motifs) influence sequencing success, we studied three major sequencing technologies (Illumina, Pacific Biosciences HiFi, and Oxford Nanopore Technologies, ONT). We estimated sequencing success by computing the rates of single-nucleotide, insertion, and deletion errors, as well as by evaluating mean read depth and mean base quality. Overall, all technologies exhibited altered sequencing success for most non-B motif types. Single-nucleotide error rates were generally increased for G-quadruplexes (G4s) and Z-DNA motifs in all three technologies. Illumina and PacBio HiFi deletion error rates were also increased for all non-B types except for Z-DNA motifs, while in ONT they were increased for all non-B types except for G4 motifs. Insertion error rates for non-B motifs were highly elevated in Illumina, moderately elevated in PacBio HiFi, and only slightly elevated in ONT. Using Poisson regression modeling, we evaluated how non-B DNA and other factors influence sequencing error profiles. Overall, the effect of non-B DNA on sequencing should be considered in downstream analyses, particularly in studies with limited read depth - e.g., single-cell and ancient DNA sequencing, as well as sequencing of pooled population samples. Because of different error profiles, a combination of technologies should be considered in sequencing studies of non-B motifs. ### Competing Interest Statement The authors have declared no competing interest.