Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data
Abstract Background Understanding the complex biological mechanisms of cancer patient survival using genomic and clinical data is vital, not only to develop new treatments for patients, but also to improve survival prediction. However, highly nonlinear and high-dimension, low-sample size (HDLSS) dat...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2019-12-01
|
Series: | BMC Medical Genomics |
Subjects: | |
Online Access: | https://doi.org/10.1186/s12920-019-0624-2 |
id |
doaj-30aecae4d42c4b49a2aea6df3e654e04 |
---|---|
record_format |
Article |
spelling |
doaj-30aecae4d42c4b49a2aea6df3e654e042021-04-02T17:48:54ZengBMCBMC Medical Genomics1755-87942019-12-0112S1011310.1186/s12920-019-0624-2Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical dataJie Hao0Youngsoon Kim1Tejaswini Mallavarapu2Jung Hun Oh3Mingon Kang4Department of Biostatistics, Epidemiology and Informatics, University of PennsylvaniaDepartment of Computer Science, Kennesaw State UniversityAnalytics and Data Science Institute, Kennesaw State UniversityDepartment of Medical Physics, Memorial Sloan Kettering Cancer CenterDepartment of Computer Science, University of Nevada, Las VegasAbstract Background Understanding the complex biological mechanisms of cancer patient survival using genomic and clinical data is vital, not only to develop new treatments for patients, but also to improve survival prediction. However, highly nonlinear and high-dimension, low-sample size (HDLSS) data cause computational challenges to applying conventional survival analysis. Results We propose a novel biologically interpretable pathway-based sparse deep neural network, named Cox-PASNet, which integrates high-dimensional gene expression data and clinical data on a simple neural network architecture for survival analysis. Cox-PASNet is biologically interpretable where nodes in the neural network correspond to biological genes and pathways, while capturing the nonlinear and hierarchical effects of biological pathways associated with cancer patient survival. We also propose a heuristic optimization solution to train Cox-PASNet with HDLSS data. Cox-PASNet was intensively evaluated by comparing the predictive performance of current state-of-the-art methods on glioblastoma multiforme (GBM) and ovarian serous cystadenocarcinoma (OV) cancer. In the experiments, Cox-PASNet showed out-performance, compared to the benchmarking methods. Moreover, the neural network architecture of Cox-PASNet was biologically interpreted, and several significant prognostic factors of genes and biological pathways were identified. Conclusions Cox-PASNet models biological mechanisms in the neural network by incorporating biological pathway databases and sparse coding. The neural network of Cox-PASNet can identify nonlinear and hierarchical associations of genomic and clinical data to cancer patient survival. The open-source code of Cox-PASNet in PyTorch implemented for training, evaluation, and model interpretation is available at: https://github.com/DataX-JieHao/Cox-PASNet.https://doi.org/10.1186/s12920-019-0624-2Cox-PASNetDeep neural networkSurvival analysisGlioblastoma multiformeOvarian cancer |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jie Hao Youngsoon Kim Tejaswini Mallavarapu Jung Hun Oh Mingon Kang |
spellingShingle |
Jie Hao Youngsoon Kim Tejaswini Mallavarapu Jung Hun Oh Mingon Kang Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data BMC Medical Genomics Cox-PASNet Deep neural network Survival analysis Glioblastoma multiforme Ovarian cancer |
author_facet |
Jie Hao Youngsoon Kim Tejaswini Mallavarapu Jung Hun Oh Mingon Kang |
author_sort |
Jie Hao |
title |
Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
title_short |
Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
title_full |
Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
title_fullStr |
Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
title_full_unstemmed |
Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
title_sort |
interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data |
publisher |
BMC |
series |
BMC Medical Genomics |
issn |
1755-8794 |
publishDate |
2019-12-01 |
description |
Abstract Background Understanding the complex biological mechanisms of cancer patient survival using genomic and clinical data is vital, not only to develop new treatments for patients, but also to improve survival prediction. However, highly nonlinear and high-dimension, low-sample size (HDLSS) data cause computational challenges to applying conventional survival analysis. Results We propose a novel biologically interpretable pathway-based sparse deep neural network, named Cox-PASNet, which integrates high-dimensional gene expression data and clinical data on a simple neural network architecture for survival analysis. Cox-PASNet is biologically interpretable where nodes in the neural network correspond to biological genes and pathways, while capturing the nonlinear and hierarchical effects of biological pathways associated with cancer patient survival. We also propose a heuristic optimization solution to train Cox-PASNet with HDLSS data. Cox-PASNet was intensively evaluated by comparing the predictive performance of current state-of-the-art methods on glioblastoma multiforme (GBM) and ovarian serous cystadenocarcinoma (OV) cancer. In the experiments, Cox-PASNet showed out-performance, compared to the benchmarking methods. Moreover, the neural network architecture of Cox-PASNet was biologically interpreted, and several significant prognostic factors of genes and biological pathways were identified. Conclusions Cox-PASNet models biological mechanisms in the neural network by incorporating biological pathway databases and sparse coding. The neural network of Cox-PASNet can identify nonlinear and hierarchical associations of genomic and clinical data to cancer patient survival. The open-source code of Cox-PASNet in PyTorch implemented for training, evaluation, and model interpretation is available at: https://github.com/DataX-JieHao/Cox-PASNet. |
topic |
Cox-PASNet Deep neural network Survival analysis Glioblastoma multiforme Ovarian cancer |
url |
https://doi.org/10.1186/s12920-019-0624-2 |
work_keys_str_mv |
AT jiehao interpretabledeepneuralnetworkforcancersurvivalanalysisbyintegratinggenomicandclinicaldata AT youngsoonkim interpretabledeepneuralnetworkforcancersurvivalanalysisbyintegratinggenomicandclinicaldata AT tejaswinimallavarapu interpretabledeepneuralnetworkforcancersurvivalanalysisbyintegratinggenomicandclinicaldata AT junghunoh interpretabledeepneuralnetworkforcancersurvivalanalysisbyintegratinggenomicandclinicaldata AT mingonkang interpretabledeepneuralnetworkforcancersurvivalanalysisbyintegratinggenomicandclinicaldata |
_version_ |
1721553204512030720 |