Fig. 5: Precise correction of pathogenic mutations using PNLM-pcABE.

a Comparison of correction efficiencies for pathogenic mutations mediated by ABE8e and PNLM-pcABE in two stable HEK293T cell lines, including GALT c.413 C > T and OTC c.533 C > T. The heat map represents editing efficiency of A-to-G. The efficiency of ABE8e and PNLM-pcABE were determined by HTS. Heatmap reflects averaged data from three biological replicates. b The target sequences of the two gene pathogenicity locus in Fig. 5a. Above the sequences are the corresponding amino acids. The green sequence area represents the main editing window of ABE8e. The disease locus A is in bold red and labeled ▲. The potential pathogenicity locus is in bold blue and labeled×. The PAM sequence is in purple. Specific locus information and Variation IDs are listed. Below are alleles and frequencies of the pathogenic mutations corrected by ABE8e, ABE9 and PNLM-pcABE. The wild-type allele frequencies were omitted. c The normalized ratio of A-to-G base editing efficiency of ABE8e, ABE9 and PNLM-pcABE in the two previously mentioned stable HEK293T cell lines. Data are mean ± s.d. (n  =  3 independent experiments) d The pie chart on the top left shows distribution ratio of pathogenetic point mutations that could potentially be corrected by base editors in the ClinVar database (accessed 16 July 2024). On the top right is a schematic diagram of pathogenic point mutations correctable by ABEs. Venn diagrams on the bottom show the pathogenic mutations that can be suitably corrected by ABE8e, ABE9 and PNLM-pcABE in NGG and NG PAM contexts without introducing bystander editing (ABE8e: Corrects pathogenic A in the 3–9 editing window without additional As; ABE9: Corrects pathogenic A at position 5, other As allowed in positions 3–9 except position 6; PNLM-pcABE: Corrects pathogenic A at position 6 or 7, other As allowed in positions 3–9 except position 5). Source data are provided as a Source data file.