{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T01:41:57Z","timestamp":1760060517690,"version":"build-2065373602"},"reference-count":34,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T00:00:00Z","timestamp":1756944000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Algorithms"],"abstract":"<jats:p>CRISPR-Cas9 has emerged as a remarkably powerful gene editing tool and has advanced both research and gene therapy applications. Machine learning models have been developed to predict off-target cleavages. Despite progress, accuracy, stability, and interpretability remain open challenges. Combining predictive modeling with interpretability can provide valuable insights into model behavior and increase its trustworthiness. This study proposes a group-wise L1 regularization method guided by SHAP values. For the implementation of this method, the CRISPR-M model was used, and SHAP-informed regularization strengths were calculated and applied to features grouped by relevance. Models were trained on HEK293T and evaluated on K562. In addition to the CRISPR-M baseline, three variants were developed: L1-Grouped-Epigenetics, L1-Grouped-Complete, and L1-Uniform-Epigenetics (control). L1-Grouped-Epigenetics, using penalties split by on- and off-target epigenetic factors, moderately improved mean precision, AUPRC, and AUROC relative to the baseline, as well as showing reduced variability in precision and AUPRC across seeds, although its mean recall and F-metrics were slightly lower than those of CRISPR-M. L1-Grouped-Complete achieved the highest mean AUROC and Spearman correlation and presented lower variability than CRISPR-M for recall, F1, and F-beta, despite reduced recall and F-metrics relative to CRISPR-M. Overall, this approach required only minor architectural adjustments, making it adaptable to other models and domains. While results demonstrate potential for enhancing interpretability and robustness without sacrificing predictive performance, further validation across additional datasets is required.<\/jats:p>","DOI":"10.3390\/a18090561","type":"journal-article","created":{"date-parts":[[2025,9,5]],"date-time":"2025-09-05T07:46:17Z","timestamp":1757058377000},"page":"561","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A SHAP-Guided Grouped L1 Regularization Method for CRISPR-Cas9 Off-Target Predictions"],"prefix":"10.3390","volume":"18","author":[{"given":"Evmorfia","family":"Tentsidou","sequence":"first","affiliation":[{"name":"Interdisciplinary PSP Bioinformatics and Neuroinformatics, Hellenic Open University, 263 31 Patra, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9917-4486","authenticated-orcid":false,"given":"Haridimos","family":"Kondylakis","sequence":"additional","affiliation":[{"name":"Interdisciplinary PSP Bioinformatics and Neuroinformatics, Hellenic Open University, 263 31 Patra, Greece"},{"name":"Computer Science Department, University of Crete, Voutes Campus, 700 13 Heraklion, Greece"},{"name":"Institute of Computer Science, Foundation for Research and Technology Hellas, N. Plastira 100, 700 13 Heraklion, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,9,4]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"e184","DOI":"10.52225\/narra.v3i2.184","article-title":"Application of CRISPR-Cas9 Genome Editing Technology in Various Fields: A Review","volume":"3","author":"Ansori","year":"2023","journal-title":"Narra J."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1262","DOI":"10.1016\/j.cell.2014.05.010","article-title":"Development and Applications of CRISPR-Cas9 for Genome Engineering","volume":"157","author":"Hsu","year":"2014","journal-title":"Cell"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"777","DOI":"10.1134\/S0006297922080090","article-title":"CRISPR\u2013Cas9: A History of Its Discovery and Ethical Considerations of Its Use in Genome Editing","volume":"87","author":"Gostimskaya","year":"2022","journal-title":"Biochem. Mosc."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Mak, J.K., St\u00f6rtz, F., and Minary, P. (2022). Comprehensive Computational Analysis of Epigenetic Descriptors Affecting CRISPR-Cas9 off-Target Activity. BMC Genom., 23.","DOI":"10.1186\/s12864-022-09012-7"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Sun, J., Guo, J., and Liu, J. (2024). CRISPR-M: Predicting sgRNA off-Target Effect Using a Multi-View Deep Learning Network. PLoS Comput. Biol., 20.","DOI":"10.1371\/journal.pcbi.1011972"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Chuai, G., Ma, H., Yan, J., Chen, M., Hong, N., Xue, D., Zhou, C., Zhu, C., Chen, K., and Duan, B. (2018). DeepCRISPR: Optimized CRISPR Guide RNA Design by Deep Learning. Genome Biol., 19.","DOI":"10.1186\/s13059-018-1459-4"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"The ENCODE Project Consortium (2004). The ENCODE (ENCyclopedia of DNA Elements) Project. Science, 306, 636\u2013640.","DOI":"10.1126\/science.1105136"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Guan, Z., and Jiang, Z. (2024). A Systematic Method for Solving Data Imbalance in CRISPR Off-Target Prediction Tasks. Comput. Biol. Med., 178.","DOI":"10.1016\/j.compbiomed.2024.108781"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Ribeiro, M.T., Singh, S., and Guestrin, C. (2016, January 13\u201317). \u201cWhy Should I Trust You?\u201d: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939778"},{"key":"ref_10","first-page":"4768","article-title":"A Unified Approach to Interpreting Model Predictions","volume":"30","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"442","DOI":"10.1016\/j.tig.2020.03.005","article-title":"Opening the Black Box: Interpretable Machine Learning for Geneticists","volume":"36","author":"Azodi","year":"2020","journal-title":"Trends Genet."},{"key":"ref_12","unstructured":"Doshi-Velez, F., and Kim, B. (2017). Towards A Rigorous Science of Interpretable Machine Learning. arXiv."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Murdoch, W.J., Singh, C., Kumbier, K., Abbasi-Asl, R., and Yu, B. (2019). Interpretable Machine Learning: Definitions, Methods, and Applications. arXiv.","DOI":"10.1073\/pnas.1900654116"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Ross, A.S., Hughes, M.C., and Doshi-Velez, F. (2017, January 19\u201325). Right for the Right Reasons: Training Differentiable Models by Constraining Their Explanations. Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, Melbourne, Australia.","DOI":"10.24963\/ijcai.2017\/371"},{"key":"ref_15","unstructured":"Rieger, L., Singh, C., Murdoch, W., and Yu, B. (2020, January 21). Interpretations Are Useful: Penalizing Explanations to Align Neural Networks with Prior Knowledge. Proceedings of the 37th International Conference on Machine Learning, PMLR 119, Online."},{"key":"ref_16","first-page":"118701","article-title":"Improving Generalization and Convergence by Enhancing Implicit Regularization","volume":"37","author":"Wang","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_17","unstructured":"Cheng, Q., Qu, S., and Lee, J. (2023). SHAPNN: Shapley Value Regularized Tabular Neural Network. arXiv."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Graham, J., and Sheng, V.S. (2025). SHAP Informed Neural Network. Mathematics, 13.","DOI":"10.3390\/math13132152"},{"key":"ref_19","unstructured":"Yerramilli, S., Tamarapalli, J.S., Francis, J., and Nyberg, E. (2024). Attribution Regularization for Multimodal Paradigms. arXiv."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"620","DOI":"10.1038\/s42256-021-00343-w","article-title":"Improving Performance of Deep Learning Models with Axiomatic Attribution Priors and Expected Gradients","volume":"3","author":"Erion","year":"2021","journal-title":"Nat. Mach. Intell."},{"key":"ref_21","unstructured":"Vukadin, D., \u0160ili\u0107, M., and Dela\u010d, G. (2025). Large Language Models as Attribution Regularizers for Efficient Model Training. arXiv."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Inecik, K., and Theis, F. (2023). scARE: Attribution Regularization for Single Cell Representation Learning. Biorxiv, 2023.","DOI":"10.1101\/2023.07.05.547784"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"12777","DOI":"10.1007\/s11042-019-08453-9","article-title":"Dropout vs. Batch Normalization: An Empirical Study of Their Impact to Deep Learning","volume":"79","author":"Garbin","year":"2020","journal-title":"Multimed. Tools Appl."},{"key":"ref_24","unstructured":"Jordan, K. (2023). On the Variance of Neural Network Training with Respect to Test Sets and Distributions. arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"120500","DOI":"10.1016\/j.ins.2024.120500","article-title":"Rolling the Dice for Better Deep Learning Performance: A Study of Randomness Techniques in Deep Neural Networks","volume":"667","author":"Altarabichi","year":"2024","journal-title":"Inf. Sci."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Han, H. (2025). Challenges of Reproducible AI in Biomedical Data Science. BMC Med. Genom., 18.","DOI":"10.1186\/s12920-024-02072-6"},{"key":"ref_27","unstructured":"Ng, A.Y. (2004, January 4\u20138). Feature Selection, L1 vs. L2 Regularization, and Rotational Invariance. Proceedings of the Twenty-First International Conference on Machine Learning\u2014ICML \u201904, Banff, AB, Canada."},{"key":"ref_28","first-page":"1929","article-title":"Dropout: A Simple Way to Prevent Neural Networks from Overfitting","volume":"15","author":"Srivastava","year":"2014","journal-title":"J. Mach. Learn. Res."},{"key":"ref_29","unstructured":"Ioffe, S., and Szegedy, C. (2015, January 7\u20139). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. Proceedings of the International Conference on Machine Learning, PMLR, Lille, France."},{"key":"ref_30","unstructured":"Ferreira, P., Titov, I., and Aziz, W. (2025). Explanation Regularisation through the Lens of Attributions. arXiv."},{"key":"ref_31","unstructured":"Nayyem, N., Rakin, A., and Wang, L. (2024). Bridging Interpretability and Robustness Using LIME-Guided Model Refinement. arXiv."},{"key":"ref_32","unstructured":"Ismail, A.A., Bravo, H.C., and Feizi, S. (2021). Improving Deep Learning Interpretability by Saliency Guided Training. arXiv."},{"key":"ref_33","unstructured":"Gururaj, S., Gr\u00fcne, L., Samek, W., Lapuschkin, S., and Weber, L. (2025). Relevance-driven Input Dropout: An Explanation-guided Regularization Technique. arXiv."},{"key":"ref_34","unstructured":"Lipton, Z.C. (2017). The Mythos of Model Interpretability. arXiv."}],"container-title":["Algorithms"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/9\/561\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:39:42Z","timestamp":1760035182000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1999-4893\/18\/9\/561"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,4]]},"references-count":34,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2025,9]]}},"alternative-id":["a18090561"],"URL":"https:\/\/doi.org\/10.3390\/a18090561","relation":{},"ISSN":["1999-4893"],"issn-type":[{"type":"electronic","value":"1999-4893"}],"subject":[],"published":{"date-parts":[[2025,9,4]]}}}