Long-read sequencing for fast and robust identification of correct genome-edited alleles: PCR-based and Cas9 capture methods

Abstract

Background Recent developments in CRISPR/Cas9 genome-editing tools have facilitated the introduction of precise alleles, including genetic intervals spanning several kilobases, directly into the embryo. However, the introduction of donor templates, via homology directed repair, can be erroneous or incomplete and these techniques often produce mosaic founder animals. Thus, newly generated alleles must be verified at the sequence level across the targeted locus. Screening for the presence of the desired mutant allele using traditional sequencing methods can be challenging due to the size of the interval to be sequenced, together with the mosaic nature of founders. Methodology/Principal findings In order to help disentangle the genetic complexity of these animals, we tested the application of Oxford Nanopore Technologies long-read sequencing at the targeted locus and found that the achievable depth of sequencing is sufficient to offset the sequencing error rate associated with the technology used to validate targeted regions of interest. We have assembled an analysis workflow that facilitates interrogating the entire length of a targeted segment in a single read, to confirm that the intended mutant sequence is present in both heterozygous animals and mosaic founders. We used this workflow to compare the output of PCR-based and Cas9 capture-based targeted sequencing for validation of edited alleles. Conclusion Targeted long-read sequencing supports in-depth characterisation of all experimental models that aim to produce knock-in or conditional alleles, including those that contain a mix of genome-edited alleles. PCR- or Cas9 capture-based modalities bring different advantages to the analysis.

Streamlining mouse genome editing by integrating AAV repair template delivery and CRISPR-Cas electroporation

“…Moreover CRISPR-Cas system, though effective, might introduce incorrect alterations alongside the intended targeting 9 , 10 . Recently, to overcome the limited length of Sanger sequencing reads (500–800bp), Oxford Nanopore technologies have been used to provide long-read sequencing covering the entire length of the targeting region from a single DNA molecule 11 . It is important to highlight that correct targeting efficiencies exhibit notable variation.…”