(B) HDR assay (6nt) (D) HDR assay (1nt) (A) Creation of disrupted EmGFP mutant (6nt) (C) Conversion of eBFP to GFP (1nt) Wild type …GTG ACC ACC TTC ACC TAC GGC… Val Thr Thr Phe Thr Tyr Gly Mutant …GTG ACC ACC TAC GGC… Val Thr Thr Tyr Gly Residue 61 62 63 64 65 66 67 (B) HDR assay (6nt) (D) HDR assay (1nt) Supplementary Figure S1. Homologous recombination assays. GFP Phase contrast Cas9 RNP Cas9/donor Cas9 RNP/donor eBFP …CTG ACC CAC GGC… Residue 64 65 66 67 Leu Thr His Gly GFP …CTG ACC TAC GGC… Leu Thr Tyr Gly eBFP Cas9 RNP Cas9/donor Cas9 RNP/donor

Supplementary Figure S1. Generation of stable cell lines for HDR assays. (A) A disrupted EmGFP HEK293 stable cell line containing deletion of “CACCTT” was generated by transfecting cells with Cas9 RNPs, followed by limiting dilution and clonal isolation. (B) HDR assays were carried out by transfecting disrupted EmGFP HEK293 cells with Cas9 RNP and donor DNA, followed by flow cytometric analysis at 48 hours post transfection. Transfections without either donor DNA (Cas9 RNP) or gRNA (Cas9/donor) were used as controls. (C) A stable HEK293FT cell expressing eBFP gene was generated using Lentiviral delivery system. A point mutation from “C” to “T” would convert His66 to Tyr66, resulting in generation of a variant of GFP. (D) HDR assays were performed by transfecting eBFP-expressing HEK293 cells with Cas9 RNP and donor DNA, followed by flow cytometric analysis to determine the percentage of GFP-positive cells at 48 hours post transfection. Transfections without either donor DNA (Cas9 RNP) or gRNA (Cas9/donor) served as controls.

Supplementary Figure S2. Optimization of delivery of Cas9 RNP in HEK293 Protocol Pst 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Pulse Voltage 1150 1400 1500 1600 1700 1100 1200 1300 1000 850 950 1050 Pulse Width 30 40 # of Pulse 1 A master mix of Cas9 RNPs were prepared and electroporated into HEK293 cells using the Neon® 24-well optimization protocol. The % Indel was determined at 48 hours post transfection

and ssDNA oligonucleotide Supplementary Figure S3. Optimization of sequential delivery of Cas9 RNPs and ssDNA oligonucleotide 147-fold induction 126-fold induction Electroporation Protocol Non-PAM PAM Non-PAM or PAM strands 0.2 0.5 0.3 µg of ss Oligonucleotide Cas9 RNPs were delivered into HEK293 cells and then washed once with Resuspension Buffer R. The cell pellets were resuspended in Buffer R and supplemented with ssDNA oligonucleotide, followed by electroporation using the Neon® 24-well optimization protocol. The percentages of EmGFP-positive cells were determined by flow cytometry. Cas9 RNPs (cas9 nuclease and gRNA) and Cas9/D (Cas9 nuclease and donor without gRNA) served as controls.

Supplementary Figure S4 Supplementary Figure S4. Optimization of sequential delivery of Cas9 RNPs and dsDNA Fragment Cas9 RNPs were delivered into HEK293 cells and then washed once with Resuspension Buffer R. The cell pellets were resuspended in Buffer R and supplemented with dsDNA DNA fragment, followed by electroporation using the the Neon® 24-well optimization protocol. The percentages of EmGFP-positive cells were determined by flow cytometry. Cas9 RNPs (cas9 nuclease and gRNA) and Cas9/D (Cas9 nuclease and donor without gRNA) served as controls.

Supplementary Figure S5 Supplementary Figure S5. Both asymmetric PAM and non-PAM ssDNA donors facilitate HDR (A) Cleavage sites (B) Asymmetric donors for +3 (C) HDR efficiency at +3 -67 +30 non-PAM30-67 3’ 5’ Insertion 3’ -48 +50 5’ non-PAM50-48 3’ -30 +67 non-PAM67-30 +3 5’ 5’ -67 +30 3’ PAM67-30 -48 +50 5’ 3’ 5’ 3’ PAM48-50 3’ 5’ -30 +67 5’ 3’ PAM30-67 -3 +5 -51 +30 3’ 5’ non-PAM30-51 PAM strand is defined as the NGG-containing strand For +3: top strand For -3, +5 : bottom strand 3’ -40 +40 5’ non-PAM40-40 3’ -30 +51 5’ non-PAM51-30 -51 +30 5’ 3’ PAM51-30 5’ -40 +40 3’ PAM40-40 5’ -30 +51 3’ PAM30-51 (D) Asymmetric donors for -3/+5 (E) HDR efficiency at -3 (F) HDR efficiency at +5 3’ -67 +30 5’ non-PAM30-67 3’ -48 +50 5’ non-PAM50-48 3’ -30 +67 5’ non-PAM67-30 5’ -67 +30 3’ PAM67-30 5’ -48 +50 3’ PAM48-50 -30 +67 5’ 3’ PAM30-67 -51 +30 3’ 5’ non-PAM30-51 3’ -40 +40 5’ non-PAM40-40 3’ -30 +51 5’ non-PAM51-30 -51 +30 5’ 3’ PAM51-30 -40 +40 5’ 3’ PAM40-40 5’ -30 +51 3’ PAM30-51

Supplementary Figure S5 Supplementary Figure S5. Both asymmetric PAM and non-PAM ssDNA donors facilitate HDR (A) Three separate gRNAs flanking the insertion site (, 0) were designed and synthesized with double-stranded breaks (DSB) occurred at position -3, +3 and +5 separately. PAM strand is defined as the NGG-containing strand. The +3 gRNA’s PAM is on the top 5’ to 3’ strand (▼), whereas the -3 and +5 gRNAs have PAMS on the bottom 3’ to 5’ strand (▲). (B,D) A series of ssDNA donors were designed with various number of nucleotides on the left arm (-) and right arm (+) of the insertion site. Both the PAM and non-PAM strands were used. (C) The Cas9 RNP (1.5 ug Cas9 nuclease, 360 ng of the +3 gRNA) and ssDNA donors (10 pmol) were sequentially delivered to disrupted EmGFP stable HEK293 cell lines. At 48 hours post transfection, the % Indel was determined by the Genomic Cleavage and Detection assay (Figure 3B), whereas the percentages of EmGFP-positive cells were determined by flow cytometry. (E,F) Same as (C) except that the -3 gRNA or +5 gRNA was used.

(A) Asymmetric PAM or Non-PAM strand ssDNA annealing Supplementary Figure S6. Both asymmetric PAM and non-PAM ssDNA donor facilitate HDR (A) Asymmetric PAM or Non-PAM strand ssDNA annealing Base substitution PAM65-35 +1 65nt 35nt PAM 3’ DSB ssDNA PAM 3’ 5’ or PAM 5’ Non-PAM Non-PAM Non-PAM 3’ 5’ 35nt 65nt 5’ 3’ Resection annealing Non-PAM65-35 PAM ssDNA is defined as the PAM-containing strand (a) (b) (B) Asymmetric donor design (C) HDR efficiencies Base substitution +1 5’ 3’ PAM 3’ 5’ Non-PAM -35 +65 5’ 3’ -50 +50 5’ 3’ -65 +35 5’ 3’ -35 +65 3’ 5’ -50 +50 3’ 5’ -65 +35 3’ 5’ -40 +40 5’ 3’ -40 +40 3’ 5’

Supplementary Figure S6 Supplementary Figure S6. Both asymmetric PAM and non-PAM ssDNA donor facilitate HDR A series of ssDNA donors were designed with various number of nucleotides on the left arm (-) and right arm (+) of the insertion site (). The PAM ssDNA oligonucleotide is defined as the NGG PAM containing strand. Both the PAM and non-PAM strands were tested. The Cas9 RNP (1.5 ug Cas9 nuclease, 360 ng eBFP gRNA) and ssDNA donors (10 pmoles) were sequentially delivered into an eBFP stable HEK293 cell line. At 48 hours post transfection, the percentages of EmGFP-positive cells were determined by flow cytometry. The bar graphs represented the averages of three individual experiments.

Supplementary Figure S7. (A) Asymmetric donor design (B) HDR efficiencies (insertion) -30 +70 3’ 5’ -70 +30 3’ 5’ -50 +50 3’ 5’ -30 +70 5’ 3’ -70 +30 5’ 3’ -50 +50 5’ 3’ % GFP cells Efficiency of symmetric and asymmetric oligos for in vivo SNP correction. (A) Illustration showing gRNA sequence targetting eBFP and various asymmetric oligos containing SNP. (B) Flow cytometer analysis of cells expressing eGFP 72 hrs post transfection. 293FT-eBFP cell line transfected with gRNA and Cas9 mRNA targeting eBFP and various 100 bp symmetric and asymmetric oligos containing SNP. Method: 293FT-eBFP cell line was used for SNP correction. 150K cells plated in a single well of 24 well tissue culture plate. For HR experiment, 10 pico moles of oligo along with 150ng gRNA and 500 ng Cas9 mRNA/well were used. Transfection was performed using Lipofectamine™ messengerMAX . 1.5 micro liter of transfection reagent was used in a single well of 24 well.

Supplemental Figure S8 A) B) An aliquot of 1 x 105 Cas9-stable iPSCs in 5 µl of Resuspension Buffer R were mixed with approximately 7 µl of Resuspension Buffer R containing 500 ng beta-actin gRNA and 10 pmoles of ssDNA donor. Cell suspension without gRNA or ssDNA donor served as controls. 10 µl of sample mixture was used for electroporation with voltage set at 1300v, pulse width set at 10 ms, and the number of pulses set at 3, respectively (Program #21). The electroporated iPSCs were transferred to a 48-well plate coated with Geltrex and filled with 0.5 ml conditioned media containing 4 ng/ml bFGF and 5 µM ROCK inhibitor Y-27632. At 48 hours post transfection, cells were washed with DPBS and then lysed in 15 µl of lysis buffer. The genomic loci were amplified using the corresponding PCR primers, followed by genomic cleavage and detection assay (A). Alternatively, the resulting PCR products were then subjected to HindIII digestion. The percentage of digestion was quantified (B).

(A) Donor configuration (B) HDR efficiency determined by sequencing Supplementary Figure S9. Insertion of a FLAG tag along with a restriction site into a genomic locus (A) Donor configuration (B) HDR efficiency determined by sequencing 30nt FLAG+EcoRI 5’ 3’ ssDNA 32nt 32nt 3’ 32-3’ 3’ 5’ 32nt 32-5’ 32nt 5’ 5’ 3’ 3’ GGN 5’ The +5 gRNA targeting the EmGFP gene was used to target the bottom 3’ to 5’ strand (▲). The resulting Cas9 RNPs and ssDNA donor or short dsDNA donors with either 5’ or 3’ single stranded overhangs were delivered to cells via sequential electroporation. The insertion efficiency of a FLAG tag was determined by EcoRI digestion and sequencing. Samples in the absence of gRNA served as controls.

Supplemental Figure S10 (A) HDR efficiency in iPSC at HPRT (C) HDR efficiencies in Cas9 stable iPSC +gRNA -gRNA P7 P21 HPRT USP12 CREBBP MDC1 (B) Mutation Pattern WT :GCATTTCTCAGTCCTAAA-CAGGGTAATGGACTGGGGCTG Clone 1:GCATTTCTCAGTCCTACA-CAGGGTAATGGACTGGGGCTG Clone 4:GCATTTCTCAGTCCTAAA----------GGACTGGGGCTG Clone 5:GCATTTCTCAGTCCT-------------GGACTGGGGCTG Clone 6:GCATTTCTCAGTCCTAAAACAGGGTAATGGACTGGGGCTG Correct HDR NHEJ

Supplementary Figure S10. Precise genome editing in iPSC Supplementary Figure S10. Precise genome editing in iPSC. (A) Cas9 RNP and a non-PAM ssDNA oligonucleotide targeting HPRT were co-delivered into iPSC via electroporation using Program 7 (1200V; 30ms;1pulse) (P7) or Program 21 (1300V; 10ms; 3pulses) (P21). Samples in the absence of donor (+gRNA) or gRNA (-gRNA) served as controls. At 48 hours post transfection, the genomic locus of HPRT was PCR-amplified. The resulting PCR fragments were subjected to either the Genomic Cleavage and Detection assay or sequencing of 96 samples. The relative percentage of wild type (wt), NHEJ, and HDR was plotted (A) based on three individual experiments. Examples of edited sequences were shown in (B) and (C). (D) Various gRNAs and ssDNA donors targeting HPRT, USP12, CREBBP, and MDC1 loci were co-delivered to stable iPSC line expressing Cas9 nuclease using Program 21. The delivery of gRNA alone or donor alone served as controls (data not shown). Upon 48 hours post transfection, the corresponding genomic loci were PCR-amplified. The resulting PCR fragments were analyzed by genomic cleavage and detection assays to determine the percentages of Indel or by sequencing to determine the relative percentage of wild type (wt), NHEJ, HDR clones.

Supplementary Figure S11. Western Blot analysis of isolated iPSC clones. M Neg C1 C2 130 kd M: marker Neg: negative control C1: clone1 C2: clone2

Time-lapse imaging for HR Supplementary Video S1. Time-lapse imaging for HR 293FT-BFP cells CRISPR+ssDNA Image taken every 2 hours Image for 3 days The Cas9 RNP (1.5 ug Cas9 nuclease, 360 ng eBFP gRNA) and ssDNA donors (10 pmoles) were transfected into an eBFP stable HEK293 cell line. The images were recorded every 2 hours for a total of 72 hours using an IncuCyte instrument.