Fig. 3: Genome editing efficiencies of predicted low performance gRNAs.

a Sketches of theoretical ‘worst-case’ spacers where the spacer is perfectly complementary to the 3′ end of the tracrRNA (roman numeral I), forms the locked hairpin with itself (roman numeral II), is perfectly complementary to the 3′ end of the crRNA (roman numeral III), or forms a perfect hairpin along the whole spacer sequence (roman numeral IV). The complementary region is colored orange and expected interactions are presented as hairpins or indicated by red arrows. b Strategy to introduce the theoretical ‘worst-case’ spacer target sequences from c into 409-B2 iCRISPR hiPSCs. Cleavage of the FRMD7 locus by Cas9 HiFi RNP is followed by subsequent HDR using a ssODN donor sequence carrying the intended target site next to a TGG PAM. M3814 is added to block NHEJ and increase HDR efficiency. Four different edits are done for target insertions I–IV followed by generation of single cell-derived cellular clones. c Genome editing efficiency of theoretical ‘worst-case’ gRNAs from a and t0 or GOLD tracrRNA. gRNAs were electroporated into Cas9 expressing 409-B2 iCRISPR hiPSCs that carry the corresponding inserted ‘worst-case’ target in the FRMD7 locus. d Genome editing efficiency of gRNAs adjacent to a disease-relevant ClinVar site and predicted to be self-complementary and to have very low cleavage efficiency (Doench 2016 and Moreno-Mateos percentile score both ≤5). Compared are t0 (normal), t0 (IDT) with proprietary chemical modifications, and GOLD tracrRNA. Cas9 RNPs were electroporated into 409-B2 hiPSCs. Independent biological replicates (n = 3) are depicted by black dots and the error bars show the SEM. Source data are provided as a Source Data file.