具有局部和全局注意力机制的图注意力网络学习单样本组学数据表征

Graph Attention Network with Local and Global Attention Mechanism to Learn Single-Sample Omic Data Representation

针对生物组学数据中基因数目远大于样本数目的高维“大p小n”问题,提出一种具有局部和全局注意力机制的图注意力网络GATOr.该模型首先在组学数据上利用Pearson相关系数计算特征之间的相关性,构建组学数据的单样本网络;然后提出一种结合局部和全局注意力机制的图注意力网络从单样本网络中学习基于图的组学特征表示,从而将组学数据的高维特性转化为低维表示.实验结果表明,GATOr与其他传统分类算法相比,在分类任务的准确率及其他指标上均取得了较优性能.

Aiming at the high-dimensional“big p small n”problem where the number of genes in biomics data(denoted as p)was far more than the number of samples(denoted as n),we proposd a graph attention network GATOr with local and global attention mechanisms.Firstly,the model used Pearson correlation coefficient to calculate the correlation between features on the omic data,and constructed a single sample network of the omic data.Secondly,we proposed a graph attention network which combined local and global attention mechanisms to learn graph-based omics feature representation from a single-sample network,thereby transforming the high-dimensional characteristics of the omics data into low-dimensional representations.The experimental results show that compared with other traditional classification algorithms,GATOr achieves better performance in classification task accuracy and other indexes.