Yang Hai, Chen Rui, Li Dongdong, Wang Zhe
Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai 200237, China.
Department of Molecular Physiology and Biophysics, Vanderbilt University, Nashville, 37240 TN, USA.
Bioinformatics. 2021 Aug 25;37(16):2231-2237. doi: 10.1093/bioinformatics/btab109.
The discovery of cancer subtyping can help explore cancer pathogenesis, determine clinical actionability in treatment, and improve patients' survival rates. However, due to the diversity and complexity of multi-omics data, it is still challenging to develop integrated clustering algorithms for tumor molecular subtyping.
We propose Subtype-GAN, a deep adversarial learning approach based on the multiple-input multiple-output neural network to model the complex omics data accurately. With the latent variables extracted from the neural network, Subtype-GAN uses consensus clustering and the Gaussian Mixture model to identify tumor samples' molecular subtypes. Compared with other state-of-the-art subtyping approaches, Subtype-GAN achieved outstanding performance on the benchmark datasets consisting of ∼4000 TCGA tumors from 10 types of cancer. We found that on the comparison dataset, the clustering scheme of Subtype-GAN is not always similar to that of the deep learning method AE but is identical to that of NEMO, MCCA, VAE and other excellent approaches. Finally, we applied Subtype-GAN to the BRCA dataset and automatically obtained the number of subtypes and the subtype labels of 1031 BRCA tumors. Through the detailed analysis, we found that the identified subtypes are clinically meaningful and show distinct patterns in the feature space, demonstrating the practicality of Subtype-GAN.
The source codes, the clustering results of Subtype-GAN across the benchmark datasets are available at https://github.com/haiyang1986/Subtype-GAN.
Supplementary data are available at Bioinformatics online.
癌症亚型的发现有助于探索癌症发病机制、确定治疗中的临床可操作性并提高患者生存率。然而，由于多组学数据的多样性和复杂性，开发用于肿瘤分子亚型的集成聚类算法仍然具有挑战性。
我们提出了Subtype-GAN，一种基于多输入多输出神经网络的深度对抗学习方法，以准确地对复杂的组学数据进行建模。利用从神经网络中提取的潜在变量，Subtype-GAN使用一致性聚类和高斯混合模型来识别肿瘤样本的分子亚型。与其他最先进的亚型分析方法相比，Subtype-GAN在由来自10种癌症的约4000个TCGA肿瘤组成的基准数据集上取得了出色的性能。我们发现，在比较数据集上，Subtype-GAN的聚类方案并不总是与深度学习方法AE的聚类方案相似，但与NEMO、MCCA、VAE等优秀方法的聚类方案相同。最后，我们将Subtype-GAN应用于BRCA数据集，并自动获得了1031个BRCA肿瘤的亚型数量和亚型标签。通过详细分析，我们发现所识别的亚型具有临床意义，并且在特征空间中呈现出明显的模式，证明了Subtype-GAN的实用性。
Subtype-GAN的源代码以及在基准数据集上的聚类结果可在https://github.com/haiyang1986/Subtype-GAN获得。
补充数据可在《生物信息学》在线获取。
