[出典] REVIEW "Predicting Mutations Generated by Cas9, Base Editing, and Prime Editing in Mammalian Cells" Weller J, Pallaseni A, Koeppel J, Parts L. CRISPR J. 2023-06-20. https://doi.org/10.1089/crispr.2023.0016 [著者所属] Wellcome Sanger Institute
本レビューでは、DNA修復機構と各種編集のモデルに利用する機械学習の基礎を紹介し、大規模な編集結果を解析するためのデータセットと手法を概観し、これまでに得られて知見を概観する。こうして構築されたモデルから得られる予測は、広範なコンテクストにおける効率的な実験を設計する基盤となる。
[構成]
Introduction: Creating, Measuring, and Predicting Mutations
CRISPR-Cas DNA editing toolkit and repair of resulting lesions
Large-scale data for learning the rules of mutation generation
Machine learning principles of prediction
Mutations from Double-Stranded Breaks
Reproducible outcomes and their sequence biases
Computational tools
Alternative break structures and structural variants
Base Edits
Sequence preferences and editing window
Prediction tools
Other types of editors and editing outcomes
Prime Editing
Determinants of prime editing efficacy
Computational tools for outcome prediction
Remaining challenges for predicting prime editing outcomes
Discussion
Conclusion
[図表リスト]
FIG. 1. CRISPR-Cas DNA editors, the lesions they generate, the main pathways that repair these, and the mutations this can lead to.
FIG. 2. A self-targeting screen to collect many mutation profiles from a single experiment.
FIG. 3. Criteria for evaluating a mutation outcome prediction model.
FIG. 4. Cas9 editing outcome observations, sequence determinants, and prediction tools.
FIG. 5. Base editing outcome frequencies and determinants.
FIG. 6. Prime editing outcome determinants and prediction tools.
Table 1. Self-targeting library delivery modalities with their advantages and drawbacks
Table 2. Tools to predict mutation outcomes in mammalian cells
コメント