Fig. 4: Identification of epegRNA and endogenous target features that influence prime-editing-induced phenotypes.
From: A benchmarked, high-efficiency prime editing platform for multiplexed dropout screening
a, Cohen’s d effect size for characteristics of epegRNA design and targeted genomic loci. All features except edit position type were evaluated without +5–6 edits. Top two stop epegRNAs with the absolute largest phenotypes at day 14 post transduction analyzed per gene. For edit location in gene, beginning, middle and end refer to editing within (0–33], (33–67] or (67–100]% of a gene, respectively. For extension length, other RTT lengths include 10, 12 and 20 nt. Cohen’s d measurements greater than 0.8 in magnitude are generally considered ‘large’. The center value denotes Cohen’s d statistic estimate and bar ranges indicate 95% confidence intervals, so that intervals including effect size 0 are not significant. All epegRNA counts (n) per feature group and P values are listed in Extended Data Fig. 5a. b, Top, schematic of epegRNA target sequence. Bottom, average growth phenotypes from StopPR screen sampled from day 14 post transduction (left) for stop epegRNAs with edits specified as indicated (right). Dark gray indicates the edited position; blue indicates positions +1–3 with respect to the Cas9(H840A) nick, light green indicates +4 (non-PAM-disrupting), dark green indicates +5–6 (PAM-disrupting) and peach indicates +7–20. Numbers of stop epegRNAs denoted. c, Replicate-averaged growth phenotypes for stop and spacer- and codon-matched synonymous epegRNAs from StopPR screen sampled from day 14 post transduction. Data points colored by density, indicated by number of neighbors. Dotted lines denote phenotype cutoffs (Z < −2). Green dots indicate strong negative growth phenotypes (Z < −5) associated with 69 synonymous epegRNAs. d, Growth phenotypes for synonymous epegRNAs (bottom plot) from StopPR screen sampled from day 14 post transduction, binned by edit position relative to exon boundaries (top schematic). Phenotypes were calculated as the average of 50 epegRNAs with the strongest negative phenotype at each position. Positions A+1 and A+2 were excluded, as fewer than 50 synonymous epegRNAs targeted those positions. Vertical lines indicate 95% confidence intervals generated for each average. The horizontal dotted line denotes phenotype cutoff (Z < −2). Splice site acceptor (AG) and donor (GT) motifs are indicated in the schematic.
