A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information

Abstract

Cancer prognosis is an essential goal for early diagnosis, biomarker selection, and medical therapy. In the past decade, deep learning has successfully solved a variety of biomedical problems. However, due to the high dimensional limitation of human cancer transcriptome data and the small number of training samples, there is still no mature deep learning-based survival analysis model that can completely solve problems in the training process like overfitting and accurate prognosis. Given these problems, we introduced a novel framework called SAVAE-Cox for survival analysis of high-dimensional transcriptome data. This model adopts a novel attention mechanism and takes full advantage of the adversarial transfer learning strategy. We trained the model on 16 types of TCGA cancer RNA-seq data sets. Experiments show that our module outperformed state-of-the-art survival analysis models such as the Cox proportional hazard model (Cox-ph), Cox-lasso, Cox-ridge, Cox-nnet, and VAECox on the concordance index. In addition, we carry out some feature analysis experiments. Based on the experimental results, we concluded that our model is helpful for revealing cancer-related genes and biological functions.

ETISTP: An Enhanced Model for Brain Tumor Identification and Survival Time Prediction

“…Moreover, the authors in [ 38 ] employed a deep learning method to diagnose Alzheimer’s disease. Moreover, SAVAE-COX [ 39 ], PubMed [ 40 ], and Page-Net [ 41 ] employ deep learning for survival time prediction. Similarly, a statistical model (SM) was proposed in [ 42 ], which efficiently predicts the survival time.…”

“…Moreover, to evaluate the classification performance, the proposed model is compared with 3DCNN [ 55 ] using five modalities, i.e., t1, t1ce, t2, Flair, and segmented. The survival prediction efficiency is evaluated in comparison with SAVAE-COX [ 39 ], CoxPH [ 40 ], PAGE-Net [ 41 ], the statistical machine learning algorithm [ 42 ], SVM [ 35 ], and the random forest classifier [ 20 ]. The following subsection details the simulation parameters used in this work.…”

“…Moreover, for the classification comparative analysis, this work used the confusion matrix for tumor classification [ 14 , 28 , 55 ]. Finally, the Concordance Index ( ) was utilized to evaluate the performance of our proposed model in terms of survival prediction, which is widely used for survival analyses in different models [ 39 , 40 ].…”

“…Successful classification of brain tumors into LGG and HGG leads to the next step, i.e., survival time prediction. To this end, we use Harrell’s Concordance Index ( ), which is widely used for survival analysis in various models [ 39 , 40 ]. The value ranges from 0 to 1, where < 0.5 indicates ineffective, whereas ≥ 0.5 indicates effective survival prediction.…”

CAT-CPI: Combining CNN and transformer to learn compound image features for predicting compound-protein interactions

“…SDNN-PPI ( Li et al, 2022 ) constructs three different ways of encoding protein sequences, and then uses a self-attention mechanism to further learn semantic relationships in the sequences for Protein-Protein Interaction (PPI). SAVAE-Cox ( Meng et al, 2022 ) adopts a novel attention mechanism and takes full advantage of the adversarial transfer learning strategy, and it works for survival analysis of high-dimensional transcriptome data. Inspired by these works, we use an image-based transformer encoder to learn the information in the images of compound molecules.…”