School of Informatics, Xiamen University, Xiamen 361001, China.
Sensors (Basel). 2021 Mar 10;21(6):1938. doi: 10.3390/s21061938.
With its increasing incidence, cancer has become one of the main causes of worldwide mortality. In this work, we mainly propose a novel attention-based neural network model named Gated Graph ATtention network (GGAT) for cancer prediction, where a gating mechanism (GM) is introduced to work with the attention mechanism (AM), to break through the previous work's limitation of 1-hop neighbourhood reasoning. In this way, our GGAT is capable of fully mining the potential correlation between related samples, helping for improving the cancer prediction accuracy. Additionally, to simplify the datasets, we propose a hybrid feature selection algorithm to strictly select gene features, which significantly reduces training time without affecting prediction accuracy. To the best of our knowledge, our proposed GGAT achieves the state-of-the-art results in cancer prediction task on LIHC, LUAD, KIRC compared to other traditional machine learning methods and neural network models, and improves the accuracy by 1% to 2% on Cora dataset, compared to the state-of-the-art graph neural network methods.
随着癌症发病率的不断上升，它已成为全球死亡率的主要原因之一。在这项工作中，我们主要提出了一种名为门控图注意力网络（GGAT）的新型基于注意力的神经网络模型，用于癌症预测，其中引入了门控机制（GM）与注意力机制（AM）配合使用，突破了之前工作中 1 跳邻域推理的局限性。这样，我们的 GGAT 能够充分挖掘相关样本之间的潜在相关性，有助于提高癌症预测的准确性。此外，为了简化数据集，我们提出了一种混合特征选择算法，严格选择基因特征，这大大减少了训练时间，而不影响预测准确性。据我们所知，与传统的机器学习方法和神经网络模型相比，我们提出的 GGAT 在 LIHC、LUAD、KIRC 上的癌症预测任务中达到了最先进的水平，并且与最先进的图神经网络方法相比，在 Cora 数据集上的准确性提高了 1%至 2%。
