A generative deep neural network for pan-digestive tract cancer survival analysis
Abstract Background The accurate identification of molecular subtypes in digestive tract cancer (DTC) is crucial for making informed treatment decisions and selecting potential biomarkers. With the rapid advancement of artificial intelligence, various machine learning algorithms have been successful...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BioData Mining |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13040-025-00426-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571956500627456 |
|---|---|
| author | Lekai Xu Tianjun Lan Yiqian Huang Liansheng Wang Junqi Lin Xinpeng Song Hui Tang Haotian Cao Hua Chai |
| author_facet | Lekai Xu Tianjun Lan Yiqian Huang Liansheng Wang Junqi Lin Xinpeng Song Hui Tang Haotian Cao Hua Chai |
| author_sort | Lekai Xu |
| collection | DOAJ |
| description | Abstract Background The accurate identification of molecular subtypes in digestive tract cancer (DTC) is crucial for making informed treatment decisions and selecting potential biomarkers. With the rapid advancement of artificial intelligence, various machine learning algorithms have been successfully applied in this field. However, the complexity and high dimensionality of the data features may lead to overlapping and ambiguous subtypes during clustering. Results In this study, we propose GDEC, a multi-task generative deep neural network designed for precise digestive tract cancer subtyping. The network optimization process involves employing an integrated loss function consisting of two modules: the generative-adversarial module facilitates spatial data distribution understanding for extracting high-quality information, while the clustering module aids in identifying disease subtypes. The experiments conducted on digestive tract cancer datasets demonstrate that GDEC exhibits exceptional performance compared to other advanced methodologies and can separate different cancer molecular subtypes that possess both statistical and biological significance. Subsequently, 21 hub genes related to pan-DTC heterogeneity and prognosis were identified based on the subtypes clustered by GDEC. The following drug analysis suggested Dasatinib and YM155 as potential therapeutic agents for improving the prognosis of patients in pan-DTC immunotherapy, thereby contributing to the enhancement of cancer patient survival. Conclusions The experiment indicate that GDEC outperforms better than other deep-learning-based methods, and the interpretable algorithm can select biologically significant genes and potential drugs for DTC treatment. |
| format | Article |
| id | doaj-art-cae0a953862b4ff6a97d4c198ceb8733 |
| institution | Kabale University |
| issn | 1756-0381 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BioData Mining |
| spelling | doaj-art-cae0a953862b4ff6a97d4c198ceb87332025-02-02T12:11:41ZengBMCBioData Mining1756-03812025-01-0118111810.1186/s13040-025-00426-zA generative deep neural network for pan-digestive tract cancer survival analysisLekai Xu0Tianjun Lan1Yiqian Huang2Liansheng Wang3Junqi Lin4Xinpeng Song5Hui Tang6Haotian Cao7Hua Chai8School of Mathematics, Foshan UniversityDepartment of Oral and Maxillofacial Surgery, Sun Yat-sen Memorial Hospital of Sun Yat-Sen UniversitySchool of Mathematics, Foshan UniversityDepartment of Oral and Maxillofacial Surgery, Sun Yat-sen Memorial Hospital of Sun Yat-Sen UniversitySchool of Mathematics, Foshan UniversitySchool of Mathematics, Foshan UniversitySchool of Mathematics, Foshan UniversityDepartment of Oral and Maxillofacial Surgery, Sun Yat-sen Memorial Hospital of Sun Yat-Sen UniversitySchool of Mathematics, Foshan UniversityAbstract Background The accurate identification of molecular subtypes in digestive tract cancer (DTC) is crucial for making informed treatment decisions and selecting potential biomarkers. With the rapid advancement of artificial intelligence, various machine learning algorithms have been successfully applied in this field. However, the complexity and high dimensionality of the data features may lead to overlapping and ambiguous subtypes during clustering. Results In this study, we propose GDEC, a multi-task generative deep neural network designed for precise digestive tract cancer subtyping. The network optimization process involves employing an integrated loss function consisting of two modules: the generative-adversarial module facilitates spatial data distribution understanding for extracting high-quality information, while the clustering module aids in identifying disease subtypes. The experiments conducted on digestive tract cancer datasets demonstrate that GDEC exhibits exceptional performance compared to other advanced methodologies and can separate different cancer molecular subtypes that possess both statistical and biological significance. Subsequently, 21 hub genes related to pan-DTC heterogeneity and prognosis were identified based on the subtypes clustered by GDEC. The following drug analysis suggested Dasatinib and YM155 as potential therapeutic agents for improving the prognosis of patients in pan-DTC immunotherapy, thereby contributing to the enhancement of cancer patient survival. Conclusions The experiment indicate that GDEC outperforms better than other deep-learning-based methods, and the interpretable algorithm can select biologically significant genes and potential drugs for DTC treatment.https://doi.org/10.1186/s13040-025-00426-zCancer subtypeTumor heterogeneitySurvival analysisPan-digestive tract cancer analysis |
| spellingShingle | Lekai Xu Tianjun Lan Yiqian Huang Liansheng Wang Junqi Lin Xinpeng Song Hui Tang Haotian Cao Hua Chai A generative deep neural network for pan-digestive tract cancer survival analysis BioData Mining Cancer subtype Tumor heterogeneity Survival analysis Pan-digestive tract cancer analysis |
| title | A generative deep neural network for pan-digestive tract cancer survival analysis |
| title_full | A generative deep neural network for pan-digestive tract cancer survival analysis |
| title_fullStr | A generative deep neural network for pan-digestive tract cancer survival analysis |
| title_full_unstemmed | A generative deep neural network for pan-digestive tract cancer survival analysis |
| title_short | A generative deep neural network for pan-digestive tract cancer survival analysis |
| title_sort | generative deep neural network for pan digestive tract cancer survival analysis |
| topic | Cancer subtype Tumor heterogeneity Survival analysis Pan-digestive tract cancer analysis |
| url | https://doi.org/10.1186/s13040-025-00426-z |
| work_keys_str_mv | AT lekaixu agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT tianjunlan agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT yiqianhuang agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT lianshengwang agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT junqilin agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT xinpengsong agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT huitang agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT haotiancao agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT huachai agenerativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT lekaixu generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT tianjunlan generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT yiqianhuang generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT lianshengwang generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT junqilin generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT xinpengsong generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT huitang generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT haotiancao generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis AT huachai generativedeepneuralnetworkforpandigestivetractcancersurvivalanalysis |