| 학회 |
한국고분자학회 |
| 학술대회 |
2018년 가을 (10/10 ~ 10/12, 경주컨벤션센터) |
| 권호 |
43권 2호 |
| 발표분야 |
의료용 고분자 부문위원회 II |
| 제목 |
Deep learning-based highly accurate prediction of CRISPR-Cpf1 activity |
| 초록 |
Cpf1 is a recently reported effector endonuclease protein of the class 2 CRISPR-Cas system. Here, we developed a method for evaluating Cpf1 activity based on target sequence composition in mammalian cells in a high-throughput manner. using this approach, we determined protospacer adjacent motif sequences of two Cpf1 nucleases, from Acidaminococcus sp. BV3L6 and Lachnospiraceae bacterium ND2006 (AsCpf1 and LbCpf1, respectively), and target sequence-dependent activity profiles of AsCpf1. Based on these data sets, we developed Seq-deepCpf1, a deep learning-based algorithm, which outperformed conventional machine learning-based algorithms. Subsequent fine-tuning of Seq-deepCpf1 using data sets of AsCpf1-induced indel frequencies at endogenous target sites with chromatin accessibility information enabled the development of DeepCpf1. We provide DeepCpf1 as a web tool, which predicts AsCpf1 activities at endogenous target sites with unprecedentedly high accuracy. |
| 저자 |
Hui Kwon Kim1, Seonwoo Min2, Myungjae Song3, Sungroh Yoon4, Hyongbum Henry Kim5
|
| 소속 |
1Department of Pharmacology, 2Yonsei Univ. College of Medicine, 3Seoul, 4Republic of Korea, 5Electrical and Computer Engineering |
| 키워드 |
Genome editing; CRISPR-Cpf1; Machine learning
|
| E-Mail |
|
