Computational Biology Department, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA.
Cell Syst. 2020 May 20;10(5):397-407.e5. doi: 10.1016/j.cels.2020.04.004.
Recent advances in ligation-free, genome-wide chromatin interaction mapping such as SPRITE and ChIA-Drop have enabled the identification of simultaneous interactions involving multiple genomic loci within the same nuclei, which are informative to delineate higher-order genome organization and gene regulation mechanisms at single-nucleus resolution. Unfortunately, computational methods for analyzing multi-way chromatin interaction data are significantly underexplored. Here we develop an algorithm, called MATCHA, based on hypergraph representation learning where multi-way chromatin interactions are represented as hyperedges. Applications to SPRITE and ChIA-Drop data suggest that MATCHA is effective to denoise the data and make predictions, which greatly enhances the data quality for analyzing the properties of multi-way chromatin interactions. MATCHA provides a promising framework to significantly improve the analysis of multi-way chromatin interaction data and has the potential to offer unique insights into higher-order chromosome organization and function. MATCHA is freely available for download here: https://github.com/ma-compbio/MATCHA.
近年来，无连接、全基因组染色质互作图谱技术（如 SPRITE 和 ChIA-Drop）的发展，使得我们能够鉴定同一核内多个基因组区域之间的同时相互作用，这些相互作用信息有助于描绘更高阶的基因组结构和单细胞分辨率下的基因调控机制。不幸的是，分析多通路染色质互作数据的计算方法还远远没有得到充分探索。在这里，我们开发了一种算法，称为 MATCHA，它基于超图表示学习，其中多通路染色质互作被表示为超边。对 SPRITE 和 ChIA-Drop 数据的应用表明，MATCHA 能够有效地对数据进行去噪和预测，这极大地提高了分析多通路染色质互作特性的数据质量。MATCHA 为显著改善多通路染色质互作数据的分析提供了一个有前景的框架，并有可能为高阶染色体结构和功能提供独特的见解。MATCHA 可在此处免费下载：https://github.com/ma-compbio/MATCHA。
