A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information
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...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Cells |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4409/11/9/1421 |
| _version_ | 1797505342199824384 |
|---|---|
| author | Xiangyu Meng Xun Wang Xudong Zhang Chaogang Zhang Zhiyuan Zhang Kuijie Zhang Shudong Wang |
| author_facet | Xiangyu Meng Xun Wang Xudong Zhang Chaogang Zhang Zhiyuan Zhang Kuijie Zhang Shudong Wang |
| author_sort | Xiangyu Meng |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-10T04:17:16Z |
| format | Article |
| id | doaj.art-b5a1733899ed48dda5ed907a73531cd6 |
| institution | Directory Open Access Journal |
| issn | 2073-4409 |
| language | English |
| last_indexed | 2024-03-10T04:17:16Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cells |
| spelling | doaj.art-b5a1733899ed48dda5ed907a73531cd62023-11-23T07:59:00ZengMDPI AGCells2073-44092022-04-01119142110.3390/cells11091421A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic InformationXiangyu Meng0Xun Wang1Xudong Zhang2Chaogang Zhang3Zhiyuan Zhang4Kuijie Zhang5Shudong Wang6College of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCollege of Computer Science and Technology, Qingdao Institute of Software, China University of Petroleum, Qingdao 266580, ChinaCancer 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.https://www.mdpi.com/2073-4409/11/9/1421deep learningsurvival analysisneural networksCox regressioncancer prognosis |
| spellingShingle | Xiangyu Meng Xun Wang Xudong Zhang Chaogang Zhang Zhiyuan Zhang Kuijie Zhang Shudong Wang A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information Cells deep learning survival analysis neural networks Cox regression cancer prognosis |
| title | A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information |
| title_full | A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information |
| title_fullStr | A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information |
| title_full_unstemmed | A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information |
| title_short | A Novel Attention-Mechanism Based Cox Survival Model by Exploiting Pan-Cancer Empirical Genomic Information |
| title_sort | novel attention mechanism based cox survival model by exploiting pan cancer empirical genomic information |
| topic | deep learning survival analysis neural networks Cox regression cancer prognosis |
| url | https://www.mdpi.com/2073-4409/11/9/1421 |
| work_keys_str_mv | AT xiangyumeng anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT xunwang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT xudongzhang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT chaogangzhang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT zhiyuanzhang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT kuijiezhang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT shudongwang anovelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT xiangyumeng novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT xunwang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT xudongzhang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT chaogangzhang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT zhiyuanzhang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT kuijiezhang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation AT shudongwang novelattentionmechanismbasedcoxsurvivalmodelbyexploitingpancancerempiricalgenomicinformation |