A densely connected framework for cancer subtype classification

Abstract Background Reliable identification of cancer subtypes is crucial for devising personalized treatment strategies. Integrating multi-omics data has proven to be an effective method for analyzing cancer subtypes. By combining molecular information across various layers, a more comprehensive un...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu Li, Denggao Zheng, Kaijie Sun, Chi Qin, Yuchen Duan, Qingqing Zhou, Yunxia Yin, Hongxing Kan, Jili Hu
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Bioinformatics
Subjects:
Online Access:https://doi.org/10.1186/s12859-025-06230-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849332254682120192
author Yu Li
Denggao Zheng
Kaijie Sun
Chi Qin
Yuchen Duan
Qingqing Zhou
Yunxia Yin
Hongxing Kan
Jili Hu
author_facet Yu Li
Denggao Zheng
Kaijie Sun
Chi Qin
Yuchen Duan
Qingqing Zhou
Yunxia Yin
Hongxing Kan
Jili Hu
author_sort Yu Li
collection DOAJ
description Abstract Background Reliable identification of cancer subtypes is crucial for devising personalized treatment strategies. Integrating multi-omics data has proven to be an effective method for analyzing cancer subtypes. By combining molecular information across various layers, a more comprehensive understanding of biological characteristics and disease mechanisms can be achieved. Results We propose DEGCN, a novel deep learning model that integrates a three-channel Variational Autoencoder (VAE) for multi-omics dimensionality reduction and a densely connected Graph Convolutional Network (GCN) for effective subtype classification. DEGCN leverages the complementary strengths of non-linear feature extraction and graph-based relational learning, enabling accurate and robust classification of renal cancer subtypes. Experimental results demonstrate that DEGCN achieves a cross-validated classification accuracy of 97.06% ± 2.04% on renal cancer data, outperforming conventional machine learning algorithms and state-of-the-art deep learning models. Moreover, its generalization ability is validated on breast and gastric cancer datasets from TCGA, with cross-validated classification accuracies of 89.82% ± 2.29% and 88.64% ± 5.24%, respectively, indicating strong cross-cancer predictive performance. Conclusion The study highlights the outstanding performance of DEGCN in heterogeneous data integration and classification accuracy, demonstrating the model’s potential in cancer subtype prediction and its application in guiding clinical treatment.
format Article
id doaj-art-8da5bd516b704b0bac71cb95d2d46a12
institution Kabale University
issn 1471-2105
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Bioinformatics
spelling doaj-art-8da5bd516b704b0bac71cb95d2d46a122025-08-20T03:46:15ZengBMCBMC Bioinformatics1471-21052025-07-0126111810.1186/s12859-025-06230-0A densely connected framework for cancer subtype classificationYu Li0Denggao Zheng1Kaijie Sun2Chi Qin3Yuchen Duan4Qingqing Zhou5Yunxia Yin6Hongxing Kan7Jili Hu8School of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineSchool of Medical Information Engineering, Anhui University of Chinese MedicineAbstract Background Reliable identification of cancer subtypes is crucial for devising personalized treatment strategies. Integrating multi-omics data has proven to be an effective method for analyzing cancer subtypes. By combining molecular information across various layers, a more comprehensive understanding of biological characteristics and disease mechanisms can be achieved. Results We propose DEGCN, a novel deep learning model that integrates a three-channel Variational Autoencoder (VAE) for multi-omics dimensionality reduction and a densely connected Graph Convolutional Network (GCN) for effective subtype classification. DEGCN leverages the complementary strengths of non-linear feature extraction and graph-based relational learning, enabling accurate and robust classification of renal cancer subtypes. Experimental results demonstrate that DEGCN achieves a cross-validated classification accuracy of 97.06% ± 2.04% on renal cancer data, outperforming conventional machine learning algorithms and state-of-the-art deep learning models. Moreover, its generalization ability is validated on breast and gastric cancer datasets from TCGA, with cross-validated classification accuracies of 89.82% ± 2.29% and 88.64% ± 5.24%, respectively, indicating strong cross-cancer predictive performance. Conclusion The study highlights the outstanding performance of DEGCN in heterogeneous data integration and classification accuracy, demonstrating the model’s potential in cancer subtype prediction and its application in guiding clinical treatment.https://doi.org/10.1186/s12859-025-06230-0Cancer subtypesMulti-omicsKidney cancerVariational autoencoderDenseNet
spellingShingle Yu Li
Denggao Zheng
Kaijie Sun
Chi Qin
Yuchen Duan
Qingqing Zhou
Yunxia Yin
Hongxing Kan
Jili Hu
A densely connected framework for cancer subtype classification
BMC Bioinformatics
Cancer subtypes
Multi-omics
Kidney cancer
Variational autoencoder
DenseNet
title A densely connected framework for cancer subtype classification
title_full A densely connected framework for cancer subtype classification
title_fullStr A densely connected framework for cancer subtype classification
title_full_unstemmed A densely connected framework for cancer subtype classification
title_short A densely connected framework for cancer subtype classification
title_sort densely connected framework for cancer subtype classification
topic Cancer subtypes
Multi-omics
Kidney cancer
Variational autoencoder
DenseNet
url https://doi.org/10.1186/s12859-025-06230-0
work_keys_str_mv AT yuli adenselyconnectedframeworkforcancersubtypeclassification
AT denggaozheng adenselyconnectedframeworkforcancersubtypeclassification
AT kaijiesun adenselyconnectedframeworkforcancersubtypeclassification
AT chiqin adenselyconnectedframeworkforcancersubtypeclassification
AT yuchenduan adenselyconnectedframeworkforcancersubtypeclassification
AT qingqingzhou adenselyconnectedframeworkforcancersubtypeclassification
AT yunxiayin adenselyconnectedframeworkforcancersubtypeclassification
AT hongxingkan adenselyconnectedframeworkforcancersubtypeclassification
AT jilihu adenselyconnectedframeworkforcancersubtypeclassification
AT yuli denselyconnectedframeworkforcancersubtypeclassification
AT denggaozheng denselyconnectedframeworkforcancersubtypeclassification
AT kaijiesun denselyconnectedframeworkforcancersubtypeclassification
AT chiqin denselyconnectedframeworkforcancersubtypeclassification
AT yuchenduan denselyconnectedframeworkforcancersubtypeclassification
AT qingqingzhou denselyconnectedframeworkforcancersubtypeclassification
AT yunxiayin denselyconnectedframeworkforcancersubtypeclassification
AT hongxingkan denselyconnectedframeworkforcancersubtypeclassification
AT jilihu denselyconnectedframeworkforcancersubtypeclassification