Extended Data Fig. 2: Analysis and validation of DNMT3A base editor scanning results.

a–d, DNMT3A base editor scanning results: (a) heatmap depicting Spearman correlations between sgRNA scores at different timepoints and for citrine⁺ or citrine⁻ cells, (b) correlation between day 9 citrine⁺ sgRNA scores and either day 3 citrine⁻ sgRNA scores (left) or day 6 citrine⁺ sgRNA scores (right), (c) day 9 citrine⁺ sgRNA scores for select versus all missense sgRNAs (active site, sgRNAs targeting residues within 5 Å of zebularine or SAH (PDB: 5YX2); near 3A-3L interface, sgRNAs targeting residues called by the InterfaceResidues.py script (https://pymolwiki.org/index.php/InterfaceResidues) as at the DNMT3A–DNMT3L interface (PDB: 5YX2) or those adjacent to these residues in the linear sequence), (d) comparison of day 9 citrine⁺ sgRNA scores for missense sgRNAs targeting annotated domains versus any non-domain region of DNMT3A. Data are the average of n = 3 replicates. For c and d, dotted lines indicate ±2 s.d. of intergenic control sgRNAs, and boxplot components are as follows: center line, median; box, interquartile range; whiskers, up to 1.5 × interquartile range per the Tukey method. e–g, Structural views of the DNMT3A MTase domain (PDB: 5YX2) highlighting residues targeted by several top enriched missense sgRNAs. Day 9 citrine⁺ sgRNA scores for the corresponding sgRNAs are printed below (where multiple sgRNAs target the same residue(s), the top sgRNA is shown). h, Full timecourse for the silencing experiment shown in Fig. 2e. Data are mean ± s.d. of n = 3 replicates, and results are representative of two independent experiments. i, Deep sequencing of cells edited with sgE756 after 9 days of dox treatment (left) or with sgG532 after 3 days of dox treatment (right), either unsorted or sorted as indicated. Plots show the percentage of aligned reads with C-to-T base edits at each indicated protospacer position or the percentage of aligned reads with indels. Sequencing was performed once (n = 1).

Source data