Zhang Hui, Li Feifei, Jia Yan, Xu Bingxiang, Zhang Yiqun, Li Xiaoli, Zhang Zhihua
CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing 100101, China.
University of Chinese Academy of Sciences, Beijing 100049, China.
Nucleic Acids Res. 2017 Dec 15;45(22):12739-12751. doi: 10.1093/nar/gkx885.
High-throughput chromosome conformation capture (3C) technologies, such as Hi-C, have made it possible to survey 3D genome structure. However, obtaining 3D profiles at kilobase resolution at low cost remains a major challenge. Therefore, we herein present an algorithm for precise identification of chromatin interaction sites at kilobase resolution from MNase-seq data, termed chromatin interaction site detector (CISD), and a CISD-based chromatin loop predictor (CISD_loop) that predicts chromatin-chromatin interactions (CCIs) from low-resolution Hi-C data. We show that the predictions of CISD and CISD_loop overlap closely with chromatin interaction analysis by paired-end tag sequencing (ChIA-PET) anchors and loops, respectively. The validity of CISD/CISD_loop was further supported by a 3C assay at about 5 kb resolution. Finally, we demonstrate that only modest amounts of MNase-seq and Hi-C data are sufficient to achieve ultrahigh resolution CCI maps. Our results suggest that CCIs may result in characteristic nucleosomes arrangement patterns flanking the interaction sites, and our algorithms may facilitate precise and systematic investigations of CCIs on a larger scale than hitherto have been possible.
高通量染色体构象捕获（3C）技术，如Hi-C，使得对三维基因组结构进行检测成为可能。然而，以低成本获得千碱基分辨率的三维图谱仍然是一项重大挑战。因此，我们在此提出一种算法，用于从微球菌核酸酶测序（MNase-seq）数据中精确识别千碱基分辨率的染色质相互作用位点，称为染色质相互作用位点检测器（CISD），以及一种基于CISD的染色质环预测器（CISD_loop），它可以从低分辨率Hi-C数据中预测染色质-染色质相互作用（CCI）。我们表明，CISD和CISD_loop的预测分别与通过配对末端标签测序（ChIA-PET）的染色质相互作用分析中的锚定和环紧密重叠。CISD/CISD_loop的有效性在约5 kb分辨率的3C分析中得到进一步支持。最后，我们证明仅需适量的MNase-seq和Hi-C数据就足以实现超高分辨率的CCI图谱。我们的结果表明，CCI可能导致相互作用位点两侧出现特征性的核小体排列模式，并且我们的算法可能有助于在比以往更大的规模上对CCI进行精确而系统的研究。
