基于Hi-c数据的酵母染色体三维结构重构

Three-dimensional structure reconstruction of yeast chromosome based on Hi-c data

通过染色体交互频率数据(Hi-c)来预测染色体三维空间结构是近年表观遗传研究热点。研究表明染色体三维空间结构在生物基因表达、调控等方面起到重要作用,对其进行三维重构是研究细胞代谢过程的基本途径。针对酵母Hi-c数据在不同染色体所呈现出的统计特征,拟合出每条染色体交互频率数据分布的数学模型,然后利用梯度上升迭代算法预测并重构其三维结构,并给出模型评估指标。实验结果表明,模型具有较高可重复性和预测精确度。

The prediction of the three-dimensional spatial structure of chromosomes by chromosome interaction frequency data (Hi-c) has been a hot topic in epigenetic studies in recent years. It has been shown that the three-dimensional spatial structure of chromosome plays an important role in gene expression and regulation. Three-dimensional reconstruction is the basic way to study the process of cell metabolism. According to the statistical characteristics of yeast Hi-c data on different chromosomes, the mathematical model of the distribution of the cross-frequency data of each chromosome was fitted, and then the gradient ascending iterative algorithm was used to predict and reconstruct its three-dimensional structure. The evaluation index of the model was established. Experimental results showed that the model has high repeatability and prediction accuracy.