Prediction of off-target activities for the end-to-end design of CRISPR guide RNAs.
Prediction of off-target activities for the end-to-end design of CRISPR guide RNAs.
复制标题
DOI：
10.1038/s41551-017-0178-6
复制
发表时间：
2018-01
影响因子：
28.1
通讯作者：
Fusi N
中科院分区：
文献类型：
--
作者：
Listgarten J;Weinstein M;Kleinstiver BP;Sousa AA;Joung JK;Crawford J;Gao K;Hoang L;Elibol M;Doench JG;Fusi N
CRISPR-Cas9系统提供了前所未有的基因组编辑能力。然而，脱靶效应会导致使用效果不佳，并且成为治疗用途开发的瓶颈。在此，我们引入了第一个基于机器学习的脱靶预测方法，产生了最先进的 CRISPR-Cas9 模型，其性能优于所有其他指南设计服务。我们的方法“Elevation”由两个相互依赖的机器学习模型组成——一个用于对各个指南-目标对进行评分，另一个将这些指南-目标分数聚合成一个整体的总结指南分数。通过系统调查，我们证明 Elevation 在这两项任务上的表现均明显优于竞争方法。此外，我们是第一个系统评估指南摘要评分问题方法的人；我们表明，最广泛使用的方法有时并不比随机方法表现更好，而高程方法始终优于随机方法，有时甚至高出一个数量级。我们还引入了一种平衡活动指南和非活动指南之间的误差的评估方法，从而封装了一系列实际用例；在整个范围内，高程始终优于其他方法。最后，由于脱靶预测的大规模和计算需求，我们开发了一种基于云的服务来进行快速检索。该服务还结合了我们之前报道的目标模型 Azimuth 来提供端到端引导设计。 （https://crispr.ml：在发布之前请将此网站视为机密）。
The CRISPR-Cas9 system provides unprecedented genome editing capabilities. However, off-target effects lead to sub-optimal usage and additionally are a bottleneck in the development of therapeutic uses. Herein, we introduce the first machine learning-based approach to off-target prediction, yielding a state-of-the-art model for CRISPR-Cas9 that outperforms all other guide design services. Our approach, Elevation, consists of two interdependent machine learning models—one for scoring individual guide-target pairs, and another which aggregates these guide-target scores into a single, overall summary guide score. Through systematic investigation, we demonstrate that Elevation performs substantially better than competing approaches on both tasks. Additionally, we are the first to systematically evaluate approaches on the guide summary score problem; we show that the most widely-used method performs no better than random at times, whereas Elevation consistently outperformed it, sometimes by an order of magnitude. We also introduce an evaluation method that balances errors between active and inactive guides, thereby encapsulating a range of practical use cases; Elevation is consistently superior to other methods across the entire range. Finally, because of the large scale and computational demands of off-target prediction, we have developed a cloud-based service for quick retrieval. This service provides end-to-end guide design by also incorporating our previously reported on-target model, Azimuth. (https://crispr.ml:please treat this web site as confidential until publication).
登录
查看更多内容
影响因子：
7.3
作者：
Housden BE;Valvezan AJ;Kelley C;Sopko R;Hu Y;Roesel C;Lin S;Buckner M;Tao R;Yilmazel B;Mohr SE;Manning BD;Perrimon N
通讯作者：
Perrimon N
影响因子：
46.9
作者：
Kleinstiver BP;Tsai SQ;Prew MS;Nguyen NT;Welch MM;Lopez JM;McCaw ZR;Aryee MJ;Joung JK
通讯作者：
Joung JK
影响因子：
46.9
作者：
Doench JG;Fusi N;Sullender M;Hegde M;Vaimberg EW;Donovan KF;Smith I;Tothova Z;Wilen C;Orchard R;Virgin HW;Listgarten J;Root DE
通讯作者：
Root DE
影响因子：
48
作者：
Crosetto, Nicola;Mitra, Abhishek;Silva, Maria Joao;Bienko, Magda;Dojer, Norbert;Wang, Qi;Karaca, Elif;Chiarle, Roberto;Skrzypczak, Magdalena;Ginalski, Krzysztof;Pasero, Philippe;Rowicka, Maga;Dikic, Ivan
通讯作者：
Dikic, Ivan
影响因子：
0.5
作者：
BaezaYates, RA;Perleberg, CH
通讯作者：
Perleberg, CH