DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning

被引：289

作者：

Angermueller, Christof ^{[1
]}

Lee, Heather J. ^{[2
,3
]}

Reik, Wolf ^{[2
,3
]}

Stegle, Oliver ^{[1
]}

机构：

[1] European Bioinformat Inst, European Mol Biol Lab, Wellcome Genome Campus, Cambridge CB10 1SD, England

[2] Babraham Inst, Epigenet Programme, Cambridge, England

[3] Wellcome Trust Sanger Inst, Wellcome Genome Campus, Cambridge CB10 1SA, England

来源：

GENOME BIOLOGY | 2017年 / 18卷

基金：

英国惠康基金; 英国生物技术与生命科学研究理事会;

关键词：

Deep learning; Artificial neural network; Machine learning; Single-cell genomics; DNA methylation; Epigenetics; EMBRYONIC STEM-CELLS; SEQUENCE; SITES; DIFFERENTIATION; HEXANUCLEOTIDE; VERTEBRATE; LANDSCAPES; INFERENCE; CHROMATIN; VARIANTS;

D O I：

10.1186/s13059-017-1189-z

中图分类号：

Q81 [生物工程学（生物技术）]; Q93 [微生物学];

学科分类号：

071005 ; 0836 ; 090102 ; 100705 ;

摘要：

Recent technological advances have enabled DNA methylation to be assayed at single-cell resolution. However, current protocols are limited by incomplete CpG coverage and hence methods to predict missing methylation states are critical to enable genome-wide analyses. We report DeepCpG, a computational approach based on deep neural networks to predict methylation states in single cells. We evaluate DeepCpG on single-cell methylation data from five cell types generated using alternative sequencing protocols. DeepCpG yields substantially more accurate predictions than previous methods. Additionally, we show that the model parameters can be interpreted, thereby providing insights into how sequence composition affects methylation variability.

引用

页数：13

共 72 条

[1] Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning [J].