Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma

BackgroundOne approach to improving the personalized treatment of cancer is to understand the cellular signaling transduction pathways that cause cancer at the level of the individual patient. In this study, we used unsupervised deep learning to learn the hierarchical structure within cancer gene expression data. Deep learning is a group of machine learning algorithms that use multiple layers of hidden units to capture hierarchically related, alternative representations of the input data. We hypothesize that this hierarchical structure learned by deep learning will be related to the cellular signaling system.ResultsRobust deep learning model selection identified a network architecture that is biologically plausible. Our model selection results indicated that the 1st hidden layer of our deep learning model should contain about 1300 hidden units to most effectively capture the covariance structure of the input data. This agrees with the estimated number of human transcription factors, which is approximately 1400. This result lends support to our hypothesis that the 1st hidden layer of a deep learning model trained on gene expression data may represent signals related to transcription factor activation. Using the 3rd hidden layer representation of each tumor as learned by our unsupervised deep learning model, we performed consensus clustering on all tumor samples—leading to the discovery of clusters of glioblastoma multiforme with differential survival. One of these clusters contained all of the glioblastoma samples with G-CIMP, a known methylation phenotype driven by the IDH1 mutation and associated with favorable prognosis, suggesting that the hidden units in the 3rd hidden layer representations captured a methylation signal without explicitly using methylation data as input. We also found differentially expressed genes and well-known mutations (NF1, IDH1, EGFR) that were uniquely correlated with each of these clusters. Exploring these unique genes and mutations will allow us to further investigate the disease mechanisms underlying each of these clusters.ConclusionsIn summary, we show that a deep learning model can be trained to represent biologically and clinically meaningful abstractions of cancer gene expression data. Understanding what additional relationships these hidden layer abstractions have with the cancer cellular signaling system could have a significant impact on the understanding and treatment of cancer.

[1]  Honglak Lee,et al.  Unsupervised learning of hierarchical representations with convolutional deep belief networks , 2011, Commun. ACM.

[2]  Dong Yu,et al.  Deep Learning: Methods and Applications , 2014, Found. Trends Signal Process..

[3]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[4]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[5]  Matthew D. Wilkerson,et al.  ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking , 2010, Bioinform..

[6]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[7]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[8]  Geoffrey E. Hinton.  A Practical Guide to Training Restricted Boltzmann Machines , 2012, Neural Networks: Tricks of the Trade.

[9]  Serdar Bozdag,et al.  G-Cimp Status Prediction Of Glioblastoma Samples Using mRNA Expression Data , 2012, PloS one.

[10]  George Michailidis,et al.  Critical limitations of consensus clustering in class discovery , 2014, Scientific Reports.

[11]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[12]  Yoshua Bengio,et al.  Practical Recommendations for Gradient-Based Training of Deep Architectures , 2012, Neural Networks: Tricks of the Trade.

[13]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[14]  Marc'Aurelio Ranzato,et al.  Building high-level features using large scale unsupervised learning , 2011, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[15]  D. Haussler,et al.  The Somatic Genomic Landscape of Glioblastoma , 2013, Cell.

[16]  Rasool Fakoor,et al.  Using deep learning to enhance cancer diagnosis and classication , 2013 .

[17]  Juan M. Vaquerizas,et al.  A census of human transcription factors: function, expression and evolution , 2009, Nature Reviews Genetics.

[18]  G. Getz,et al.  GISTIC2.0 facilitates sensitive and confident localization of the targets of focal somatic copy-number alteration in human cancers , 2011, Genome Biology.

[19]  John Salvatier,et al.  Theano: A Python framework for fast computation of mathematical expressions , 2016, ArXiv.

[20]  S. Gabriel,et al.  Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. , 2010, Cancer cell.

[21]  Xinghua Lu,et al.  Trans-species learning of cellular signaling systems with bimodal deep belief networks , 2015, Bioinform..

[22]  Xinghua Lu,et al.  Learning a hierarchical representation of the yeast transcriptomic machinery using an autoencoder model , 2016, BMC Bioinformatics.

[23]  T. Hampton,et al.  The Cancer Genome Atlas , 2020, Indian Journal of Medical and Paediatric Oncology.

[24]  Honglak Lee,et al.  Sparse deep belief net model for visual area V2 , 2007, NIPS.

[25]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[26]  R. Wilson,et al.  Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. , 2010, Cancer cell.

[27]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[28]  Benjamin J. Raphael,et al.  Multiplatform Analysis of 12 Cancer Types Reveals Molecular Classification within and across Tissues of Origin , 2014, Cell.

[29]  Ting Chen,et al.  Integrative Data Analysis of Multi-Platform Cancer Data with a Multimodal Deep Learning Approach , 2015, IEEE/ACM Transactions on Computational Biology and Bioinformatics.