Fig. 1: Overall evolution-guided engineering strategy and identification of OrufIscB as an active human genome editor.

a, Comparison of IscB and Cas9 protein size, domain architecture and ncRNAs. P, PLMP domain; B, bridge helix domain; I, RuvC domain region I; II, RuvC domain region II; III, RuvC domain region III; TI, TAM-interacting domain; PI, PAM-interacting domain; ncRNA, noncoding RNA. b, Schematic overview of the overall discovery and engineering pipeline used to develop an IscB suitable for in vivo mammalian applications. c, Phylogenetic tree of all experimentally characterized IscBs and select type II-D Cas9s. Protein and associated ωRNA length, average indel activity across 12 targeted sites in the human genome in a pooled-guide assay, and dominant TAM preference are shown on the outer rings. The circled tree nodes indicate specific orthologs; red filled circles indicate the ortholog had genome editing activity in human cells. Example AlphaFold2 structures of REC linkers and mini-RECs of select orthologs are shown on the left and right, with the solved cryo-EM structure used for OgeuIscB (PDB: 7XHT (ref. ¹⁶)). d, Validation of the 12-guide pooled screen using individual guides. Indels are shown only for those combinations of protein and guide that showed activity in the pooled screen. T, targeting; NT, nontargeting control. Data are presented as mean ± s.d.; n = 4 replicate transfections. *P < 0.05; **P < 0.01; ***P < 0.001 from two-sided t-test relative to the NT guide. The P values are included in Supplementary Table 2. e, Indel formation at CA2 mediated by OrufIscB or SpCas9 using guide lengths ranging from 12 to 28 nt. The black arrows show the shortest lengths with detectable activity. Data are presented as mean ± s.d.; n = 3 replicate transfections.