{"id":598457,"date":"2023-01-17T12:57:12","date_gmt":"2023-01-17T18:57:12","guid":{"rendered":"https:\/\/news.sellorbuyhomefast.com\/index.php\/2023\/01\/17\/predicting-prime-editing-efficiency-and-product-purity-by-deep-learning\/"},"modified":"2023-01-17T12:57:12","modified_gmt":"2023-01-17T18:57:12","slug":"predicting-prime-editing-efficiency-and-product-purity-by-deep-learning","status":"publish","type":"post","link":"https:\/\/newsycanuse.com\/index.php\/2023\/01\/17\/predicting-prime-editing-efficiency-and-product-purity-by-deep-learning\/","title":{"rendered":"Predicting prime editing efficiency and product purity by deep learning"},"content":{"rendered":"\n
\n
\n

Data availability<\/h2>\n

Measured editing rates used for analysis and figures in this study are provided as Supplementary Tables and on GitHub (https:\/\/github.com\/uzh-dqbm-cmi\/PRIDICT<\/a>). DNA-sequencing data is available via the National Center for Biotechnology Information Sequence Read Archive (PRJNA825584<\/a>). Target sequences of pathogenic mutations were based on the ClinVar database (accessed December 2019), and corresponding genomic sequences (flanking the edit) were acquired via UCSC Genome Browser (Table Browser, hg38). Plasmid encoding for pCMV-PE2-tagRFP-BleoR is available from Addgene (no. 192508).<\/p>\n<\/div>\n

\n

Code availability<\/h2>\n

Custom Python code used in this study is provided on GitHub (https:\/\/github.com\/uzh-dqbm-cmi\/PRIDICT<\/a>). Additional information on the PRIDICT algorithm can be found in Supplementary Methods 1<\/a>.<\/p>\n<\/div>\n

\n

References<\/h2>\n
\n
    \n
  1. \n

    Anzalone, A. V. et al. Search-and-replace genome editing without double-strand breaks or donor DNA. Nature<\/i> 576<\/b>, 149\u2013157 (2019).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  2. \n

    Hsu, J. Y. et al. PrimeDesign software for rapid and simplified design of prime editing guide RNAs. Nat. Commun.<\/i> 12<\/b>, 1034 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  3. \n

    Hwang, G.-H. et al. PE-Designer and PE-Analyzer: web-based design and analysis tools for CRISPR prime editing. Nucleic Acids Res.<\/i> 49<\/b>, W499\u2013W504 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  4. \n

    Kim, H. K. et al. Predicting the efficiency of prime editing guide RNAs in human cells. Nat. Biotechnol.<\/i> 39<\/b>, 198\u2013206 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  5. \n

    Li, Y., Chen, J., Tsai, S. Q. & Cheng, Y. Easy-Prime: a machine learning\u2013based prime editor design tool. Genome Biol.<\/i> 22<\/b>, 235 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  6. \n

    Landrum, M. J. et al. ClinVar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res.<\/i> 46<\/b>, D1062\u2013D1067 (2018).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  7. \n

    Nielsen, S., Yuzenkova, Y. & Zenkin, N. Mechanism of eukaryotic RNA polymerase III transcription termination. Science<\/i> 340<\/b>, 1577\u20131580 (2013).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  8. \n

    Gao, Z., Herrera-Carrillo, E. & Berkhout, B. Delineation of the exact transcription termination signal for type 3 polymerase III. Mol. Ther. Nucleic Acids<\/i> 10<\/b>, 36\u201344 (2018).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  9. \n

    Bill, C. A., Duran, W. A., Miselis, N. R. & Nickoloff, J. A. Efficient repair of all types of single-base mismatches in recombination intermediates in Chinese hamster ovary cells: competition between long-patch and G-T glycosylase-mediated repair of G-T mismatches. Genetics<\/i> 149<\/b>, 1935\u20131943 (1998).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  10. \n

    Walton, R. T., Christie, K. A., Whittaker, M. N. & Kleinstiver, B. P. Unconstrained genome targeting with near-PAMless engineered CRISPR-Cas9 variants. Science<\/i> 368<\/b>, 290\u2013296 (2020).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  11. \n

    Lundberg, S. M. & Lee, S. I. A unified approach to interpreting model predictions. In Proc. 31st International Conference on Neural Information Processing Systems<\/i> (eds von Luxburg, U. et al.) 4768\u20134777 (Curran Associates Inc., 2017).<\/p>\n<\/li>\n

  12. \n

    Kim, H. K. et al. SpCas9 activity prediction by DeepSpCas9, a deep learning\u2013based model with high generalization performance. Sci. Adv.<\/i> 5<\/b>, eaax9249 (2019).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  13. \n

    Sundararajan, M., Taly, A. & Yan, Q. Axiomatic attribution for deep networks. In Proc. 34th International Conference on Machine Learning<\/i> (eds Precup, D. & Teh, Y. W.) 3319\u20133328 (PMLR, 2017).<\/p>\n<\/li>\n

  14. \n

    Doench, J. G. et al. Rational design of highly active sgRNAs for CRISPR-Cas9\u2013mediated gene inactivation. Nat. Biotechnol.<\/i> 32<\/b>, 1262\u20131267 (2014).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  15. \n

    Nelson, J. W. et al. Engineered pegRNAs improve prime editing efficiency. Nat. Biotechnol.<\/i> 40<\/b>, 402\u2013410 (2022).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  16. \n

    Chen, P. J. et al. Enhanced prime editing systems by manipulating cellular determinants of editing outcomes. Cell<\/i> 184<\/b>, 5635\u20135652.e29 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  17. \n

    Nair, N. et al. Computationally designed liver-specific transcriptional modules and hyperactive factor IX improve hepatic gene therapy. Blood<\/i> 123<\/b>, 3195\u20133199 (2014).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  18. \n

    Untergasser, A. et al. Primer3\u2014new capabilities and interfaces. Nucleic Acids Res.<\/i> 40<\/b>, e115 (2012).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  19. \n

    Villiger, L. et al. Treatment of a metabolic liver disease by in vivo genome base editing in adult mice. Nat. Med<\/i>. 24<\/b>, 1519\u20131525 (2018).<\/p>\n<\/li>\n

  20. \n

    Kim, H. K. et al. In vivo high-throughput profiling of CRISPR-Cpf1 activity. Nat. Methods<\/i> 14<\/b>, 153\u2013159 (2017).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  21. \n

    Kent, W. J. et al. The human genome browser at UCSC. Genome Res.<\/i> 12<\/b>, 996\u20131006 (2002).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  22. \n

    Kim, N. et al. Prediction of the sequence-specific cleavage activity of Cas9 variants. Nat. Biotechnol.<\/i> 38<\/b>, 1328\u20131336 (2020).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  23. \n

    Dang, Y. et al. Optimizing sgRNA structure to improve CRISPR-Cas9 knockout efficiency. Genome Biol.<\/i> 16<\/b>, 280 (2015).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  24. \n

    B\u00f6ck, D. et al. In vivo prime editing of a metabolic liver disease in mice. Sci. Transl. Med.<\/i> 14<\/b>, eabl9238 (2022).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  25. \n

    Jensen, K. T. et al. Chromatin accessibility and guide sequence secondary structure affect CRISPR-Cas9 gene editing efficiency. FEBS Lett.<\/i> 591<\/b>, 1892\u20131901 (2017).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  26. \n

    Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J.<\/i> 17<\/b>, 10 (2011).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  27. \n

    Shen, W., Le, S., Li, Y. & Hu, F. SeqKit: a cross-platform and ultrafast toolkit for FASTA\/Q file manipulation. PLoS ONE<\/i> 11<\/b>, e0163962 (2016).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  28. \n

    Lorenz, R. et al. ViennaRNA package 2.0. Algorithms Mol. Biol.<\/i> 6<\/b>, 26 (2011).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  29. \n

    Clement, K. et al. CRISPResso2 provides accurate and rapid genome editing sequence analysis. Nat. Biotechnol.<\/i> 37<\/b>, 224\u2013226 (2019).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  30. \n

    Schep, R. et al. Impact of chromatin context on Cas9-induced DNA double-strand break repair pathway balance. Mol. Cell<\/i> 81<\/b>, 2216\u20132230.e10 (2021).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  31. \n

    Barrett, T. et al. NCBI GEO: archive for functional genomics data sets\u2014update. Nucleic Acids Res.<\/i> 41<\/b>, D991\u2013D995 (2012).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  32. \n

    Luo, Y. et al. New developments on the Encyclopedia of DNA Elements (ENCODE) data portal. Nucleic Acids Res.<\/i> 48<\/b>, D882\u2013D889 (2020).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  33. \n

    Karabacak Calviello, A., Hirsekorn, A., Wurmus, R., Yusuf, D. & Ohler, U. Reproducible inference of transcription factor footprints in ATAC-seq and DNase-seq datasets using protocol-specific bias modeling. Genome Biol.<\/i> 20<\/b>, 42 (2019).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  34. \n

    Lamb, K. N. et al. Discovery and characterization of a cellular potent positive allosteric modulator of the polycomb repressive complex 1 chromodomain, CBX7. Cell Chem. Biol.<\/i> 26<\/b>, 1365\u20131379.e22 (2019).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  35. \n

    Hattori, T. et al. Antigen clasping by two antigen-binding sites of an exceptionally specific antibody for histone methylation. Proc. Natl Acad. Sci. USA<\/i> 113<\/b>, 2092\u20132097 (2016).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  36. \n

    Lee, B. T. et al. The UCSC Genome Browser database: 2022 update. Nucleic Acids Res.<\/i> 50<\/b>, D1115\u2013D1122 (2022).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  37. \n

    Zerbino, D. R., Johnson, N., Juettemann, T., Wilder, S. P. & Flicek, P. WiggleTools: parallel processing of large collections of genome-wide datasets for visualization and statistical analysis. Bioinformatics<\/i> 30<\/b>, 1008\u20131009 (2014).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  38. \n

    Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res.<\/i> 12<\/b>, 2825\u20132830 (2011).<\/p>\n


    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  39. \n

    Chen, T. & Guestrin, C. XGBoost: a scalable tree boosting system. In Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining<\/i> (eds Krishnapuram, B. et al.) 785\u2013794 (ACM, 2016).<\/p>\n<\/li>\n

  40. \n

    Marquart, K. F. et al. Predicting base editing outcomes with an attention-based deep learning algorithm trained on high-throughput target library screens. Nat. Commun.<\/i> 12<\/b>, 1\u201325 (2020).<\/p>\n


    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  41. \n

    Paszke, A. et al. Automatic differentiation in pytorch. In Proc. 31st Annual Conference on Neural Information Processing Systems:Advances in Neural Information Processing Systems 2017<\/i> (NIPS, 2017).<\/p>\n<\/li>\n

  42. \n

    Cho, K. et al. Learning phrase representations using RNN encoder\u2013decoder for statistical machine translation. In Proc. 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)<\/i> (eds Moschitti, A. et al.) 1724\u20131734 (Association for Computational Linguistics, 2014).<\/p>\n<\/li>\n

  43. \n

    Chung, J., Gulcehre, C., Cho, K. & Bengio, Y. Empirical evaluation of gated recurrent neural networks on sequence modeling. Preprint at https:\/\/arxiv.org\/abs\/1412.3555<\/a> (2014).<\/p>\n<\/li>\n

  44. \n

    Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Comput.<\/i> 9<\/b>, 1735\u20131780 (1997).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  45. \n

    Bengio, Y., Simard, P. & Frasconi, P. Learning long-term dependencies with gradient descent is difficult. IEEE Trans. Neural Netw.<\/i> 5<\/b>, 157\u2013166 (1994).<\/p>\n

    Article<\/a>\u00a0
    \n     CAS<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  46. \n

    Graves, A. Supervised Sequence Labelling with Recurrent Neural Networks<\/i> 385 (Springer, 2012).<\/p>\n<\/li>\n

  47. \n

    Luong, T., Pham, H. & Manning, C. D. Effective approaches to attention-based neural machine translation. In Proc. 2015 Conference on Empirical Methods in Natural Language Processing<\/i> (eds M\u00e0rquez, L. et al.) 1412\u20131421 (Association for Computational Linguistics, 2015).<\/p>\n<\/li>\n

  48. \n

    Vaswani, A. et al. Attention is all you need. In Proc. 31st International Conference on Neural Information Processing Systems<\/i> (eds von Luxburg, U. et al.) 6000\u20136010 (Curan Associates Inc., 2017).<\/p>\n<\/li>\n

  49. \n

    Ba, J. L., Kiros, J. R. & Hinton, G. E. Layer normalization. Preprint at https:\/\/arxiv.org\/abs\/1607.06450<\/a> (2016).<\/p>\n<\/li>\n

  50. \n

    He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proc. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)<\/i> 770\u2013778 (IEEE, 2016).<\/p>\n<\/li>\n

  51. \n

    Bergstra, J. & Bengio, Y. Random search for hyper-parameter optimization. J. Mach. Learn. Res.<\/i> 13<\/b>, 281\u2013305 (2012).<\/p>\n


    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n

  52. \n

    Eggington, J. M., Greene, T. & Bass, B. L. Predicting sites of ADAR editing in double-stranded RNA. Nat. Commun.<\/i> 2<\/b>, 319 (2011).<\/p>\n

    Article<\/a>\u00a0
    \n
    \n Google Scholar<\/a>\u00a0\n <\/p>\n<\/li>\n<\/ol>\n

    Download references<\/a><\/p>\n<\/div>\n<\/div>\n

    \n

    Acknowledgements<\/h2>\n

    We thank the Functional Genomics Center Zurich for their help and support in next-generation sequencing; the Flow Cytometry Facility of the University of Zurich and especially M. Wickert for performing liver hepatocyte sorting experiments; the Science IT team at the University of Zurich for providing infrastructure used for data analysis and especially P. Shemella for helpful discussions about code performance optimizations; R. Schep for discussions about chromatin marks; G. Affentranger for support in the design of figures; the members of the Schwank laboratory for fruitful discussions. This work was supported by the SNF (grant nos. 310030_185293 and 201184), the University Research Priority Program \u2018Human Reproduction Reloaded\u2019 and \u2018ITINERARE\u2019 of the University of Zurich. K.F.M. holds a PHRT iDoc Fellowship (PHRT_324).<\/p>\n<\/div>\n

    \n

    Author information<\/h2>\n
    \n

    Author notes<\/span><\/p>\n

      \n
    1. \n

      These authors contributed equally: Nicolas Mathis, Ahmed Allam.<\/p>\n<\/li>\n<\/ol>\n

      Authors and Affiliations<\/h3>\n
        \n
      1. \n

        Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland<\/p>\n

        Nicolas Mathis,\u00a0Lucas Kissling,\u00a0Kim Fabiano Marquart,\u00a0Lukas Schmidheini,\u00a0Cristina Solari\u00a0&\u00a0Gerald Schwank<\/p>\n<\/li>\n

      2. \n

        Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland<\/p>\n

        Ahmed Allam,\u00a0Zsolt Bal\u00e1zs\u00a0&\u00a0Michael Krauthammer<\/p>\n<\/li>\n

      3. \n

        Institute of Molecular Health Sciences, ETH Zurich, Zurich, Switzerland<\/p>\n

        Kim Fabiano Marquart\u00a0&\u00a0Lukas Schmidheini<\/p>\n<\/li>\n<\/ol>\n

        Contributions<\/h3>\n

        N.M. designed the study, performed experiments and analyzed data. A.A. designed and generated attention-based bidirectional RNNs (PRIDICT) and implemented feature extraction strategies. A.A. and N.M. built linear regression and tree-based machine learning models and performed feature extraction analysis. L.K. performed in vivo experiments. K.F.M. and C.S. contributed to arrayed validation experiments. L.S. performed pegRNA and AdV cloning experiments. Z.B. performed the analysis of chromatin characteristics of endogenous loci. N.M., A.A. and G.S. wrote the manuscript. M.K. and G.S. designed and supervised the research. All authors revised the manuscript.<\/p>\n

        Corresponding authors<\/h3>\n

        Correspondence to
        \n Michael Krauthammer<\/a> or Gerald Schwank<\/a>.<\/p>\n<\/div>\n<\/div>\n

        \n

        Ethics declarations<\/h2>\n
        \n

        Competing interests<\/h3>\n

        The authors declare no competing interests.<\/p>\n<\/p><\/div>\n<\/div>\n

        \n

        Peer review<\/h2>\n
        \n

        Peer review information<\/h3>\n

        Nature Biotechnology<\/i> thanks Sangsu Bae and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.<\/p>\n<\/p><\/div>\n<\/div>\n

        \n

        Additional information<\/h2>\n

        Publisher\u2019s note<\/b> Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.<\/p>\n<\/div>\n

        \n

        Extended data<\/h2>\n
        \n
        \n

        Extended Data Fig. 1 Self-targeting screen characteristics.<\/a><\/h3>\n

        a<\/b>, Visualization of library design (library 1) and numbers before and after filtering results. b<\/b>, Distribution of edit positions for single base replacement edits in library 1. c<\/b>, Distribution of edit positions for insertion edits in library 1. d<\/b>, Distribution of edit positions for deletion edits in library 1. e<\/b>, Distribution of insertion lengths in library 1. f<\/b>, Distribution of deletion lengths in library 1. g<\/b>, Distribution of edit types in library 1 (number of design variants and percentage of the total library). h<\/b>,i<\/b>, Editing rates of a test self-targeting locus with a forward (Fw) or reverse (Rv) orientation of the target sequence. Either on plasmid level or integrated by lentiviral transduction in HEK293T cells. Data points for bars (from left) 2,3 and 5,6 correspond to two technical replicates (simultaneous transfection of two separate wells). Only one data point was used for the plasmid controls (bar 1 and 4). h<\/b>, pegRNA with TAG to TGG edit. i<\/b>, pegRNA with TAG to TAC edit. The observed editing in the forward direction in the absence of PE2 could be caused by lentiviral reshuffling or ADAR-mediated A to I (G) RNA editing. The latter could occur during lentiviral packaging in HEK293T cells: HEK293T cells endogenously express ADAR and the target site is present as RNA on the lentiviral vector and targeted by the complementary pegRNA with a mismatch, providing an ideal template for ADAR-dependent RNA editing. The observation that primarily TAG to TGG (but not TAG to TAC) showed background editing is in line with this hypothesis, as previous studies showed ADAR preference for UAG sequences52<\/a><\/sup>.<\/p>\n<\/div>\n

        \n

        Extended Data Fig. 2 Additional validation of the DeepPE model.<\/a><\/h3>\n

        a<\/b>, Predicted (PRIDICT) and measured intended editing efficiency for GtoC edits at position 5 of RTT in the dataset from this study. Data from all five test sets (fivefold cross-validation) were combined for this visualization. n<\/i>\u2009=\u2009540. b<\/b>, Evaluation of attention-based bidirectional RNN (PRIDICT-AttnBiRNN; trained on the dataset from this study) by testing on pegRNAs from Kim et al. 2021 HT dataset (only G to C at Position 5). n<\/i>\u2009=\u20094,457. c<\/b>, Evaluation of DeepPE model (original, trained on Kim et al. 2021 HT dataset) by testing on the dataset from this study (only G to C at Position 5). n<\/i>\u2009=\u2009540. d<\/b>,e<\/b>, SHAP analysis of XGBoost models trained and tested on DeepPE dataset (n<\/i>\u2009=\u200943,149) (d<\/b>) or on G-to-C Position 5 edits from library 1 (e<\/b>). Feature descriptions are listed in Supplementary Table 1<\/a>. f<\/b>, Editing efficiency with different RTT overhang lengths (5, 7, 10, 15\u2009bp) in DeepPE (Kim et al.) dataset. n<\/i> for each bar (left to right) = 10,746, 10,828, 10,921, 10,654. Error bars = mean \u00b1s.d. g<\/b>, Editing efficiency with different RTT overhang lengths (3, 7, 10, 15\u2009bp) in GtoC Pos. 5 edits of library 1 for a direct comparison to identical edits in the DeepPE dataset. n<\/i> for each bar (left to right) = 135, 135, 137, 133. (f<\/b>,g<\/b>) Error bars = mean \u00b1s.d. h<\/b>,i<\/b>, Evaluation of DeepPE model (n<\/i>\u2009=\u200918) on 18\/45 endogenous edits from this study in HEK293T (h<\/b>) and K562 (i<\/b>).<\/p>\n<\/div>\n

        \n

        Extended Data Fig. 3 Additional validation of the Easy-Prime PE2 model.<\/a><\/h3>\n

        a<\/b>, Edit type count distribution in the original Easy-Prime test dataset. b<\/b>, Evaluation of Easy-Prime PE2 model by testing this XGBoost model on the original Easy-Prime test dataset5<\/a><\/sup>, filtered against 1\u2009bp edits at position 5 of the RTT to eliminate the bias towards this edit type. n<\/i>\u2009=\u2009585. c<\/b>\u2013g<\/b>, Evaluation of Easy-Prime PE2 by testing the model on datasets generated in this study. c<\/b>, Library 1 in HEK293T, n<\/i>\u2009=\u200992,423. d<\/b>, Library 2 (editing with PE2 and pegRNAs without tevopreQ1) in HEK293T, n<\/i>\u2009=\u2009915. e<\/b>, Library 2 (editing with PE2 and pegRNAs without tevopreQ1) in K562, n<\/i>\u2009=\u2009876. f<\/b>,g<\/b>, Endogenous loci from Fig. 4a, b<\/a> in HEK293T (f<\/b>) and K562 (g<\/b>), n<\/i>\u2009=\u200945. h<\/b>, Intended editing efficiency rank of the best-predicted pegRNA for each pathogenic locus in library 1 (PRIDICT and Easy-Prime). Pathogenic loci with multiple pegRNAs on rank 1 (identical efficiency) and loci with less than three pegRNAs were excluded from this analysis. Predictions from PRIDICT were taken from five different cross-validations to ensure none of the predictions are included in the training set. n<\/i>\u2009=\u200912,189. i<\/b>, Intended editing efficiency rank of the best-predicted pegRNA for each endogenous locus (PRIDICT and Easy-Prime). n<\/i>\u2009=\u200915.<\/p>\n<\/div>\n

        \n

        Extended Data Fig. 4 Additional library 2 evaluation with PEmax.<\/a><\/h3>\n

        a<\/b>, Mean editing efficiencies of each replicate, including all pegRNAs in library 2 with different experimental conditions in U2OS and K562 cells. Error bars indicate the mean \u00b1s.d. of three biologically independent replicates. n<\/i>\u2009=\u20093. Mean editing of library 2 for each of the three replicates is based on the following number of pegRNAs for each data point (bars left to right) = 916, 922, 917, 924, 879, 869, 877, 866. Note that absolute levels of editing efficiency for PEmax cannot be directly compared to PE2 in this study due to the use of different selection agents (Blasticidin for PEmax screens compared to Zeocin for PE2 screens). Previous studies showed that in identical setups, PEmax surpasses the performance of PE216<\/a><\/sup>. b<\/b>, Spearman correlation for PEmax editing efficiencies in library 2 between different experimental conditions (MLH1dn, tevopreQ1) and cell lines (K562, U2OS). c<\/b>, Editing efficiency rank correlations (Spearman) in library 2 between editing performed with PE2 versus editing performed with PEmax.<\/p>\n<\/div>\n<\/div>\n<\/div>\n

        \n

        Supplementary information<\/h2>\n<\/div>\n