Format

Send to

Choose Destination
Cancers (Basel). 2019 Jan 8;11(1). pii: E53. doi: 10.3390/cancers11010053.

Prognostic Gene Discovery in Glioblastoma Patients using Deep Learning.

Wong KK1,2,3, Rostomily R4, Wong STC5,6,7,8.

Author information

1
Department of Systems Medicine and Bioengineering, Houston Methodist, Houston, TX 77030, USA. KWong@houstonmethodist.org.
2
Department of Neurological Surgery, Weill Cornell Medicine, New York, NY 10065, USA. KWong@houstonmethodist.org.
3
Department of Radiology, Weill Cornell Medicine, New York, NY 10065, USA. KWong@houstonmethodist.org.
4
Department of Neurosurgery, Houston Methodist Neurological Institute, Houston, TX 77030, USA. rrostomily@houstonmethodist.org.
5
Department of Systems Medicine and Bioengineering, Houston Methodist, Houston, TX 77030, USA. stwong@houstonmethodist.org.
6
Department of Radiology, Weill Cornell Medicine, New York, NY 10065, USA. stwong@houstonmethodist.org.
7
Department of Neuroscience, Weill Cornell Medicine, New York, NY 10065, USA. stwong@houstonmethodist.org.
8
Department of Pathology and Laboratory Medicine, Weill Cornell Medicine, New York, NY 10065, USA. stwong@houstonmethodist.org.

Abstract

This study aims to discover genes with prognostic potential for glioblastoma (GBM) patients' survival in a patient group that has gone through standard of care treatments including surgeries and chemotherapies, using tumor gene expression at initial diagnosis before treatment. The Cancer Genome Atlas (TCGA) GBM gene expression data are used as inputs to build a deep multilayer perceptron network to predict patient survival risk using partial likelihood as loss function. Genes that are important to the model are identified by the input permutation method. Univariate and multivariate Cox survival models are used to assess the predictive value of deep learned features in addition to clinical, mutation, and methylation factors. The prediction performance of the deep learning method was compared to other machine learning methods including the ridge, adaptive Lasso, and elastic net Cox regression models. Twenty-seven deep-learned features are extracted through deep learning to predict overall survival. The top 10 ranked genes with the highest impact on these features are related to glioblastoma stem cells, stem cell niche environment, and treatment resistance mechanisms, including POSTN, TNR, BCAN, GAD1, TMSB15B, SCG3, PLA2G2A, NNMT, CHI3L1 and ELAVL4.

KEYWORDS:

deep learning; discovery; glioblastoma; glioblastoma stem cells; survival prediction

Supplemental Content

Full text links

Icon for Multidisciplinary Digital Publishing Institute (MDPI) Icon for PubMed Central
Loading ...
Support Center