Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma
Open accessThe DOI matches its Crossref record; 9 of the 9 references it cites were found in a registry or an index.
Abstract
BACKGROUND: One 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. RESULTS: Robust 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. CONCLUSIONS: In 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.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
The Somatic Genomic Landscape of Glioblastoma
Brennan, C. W., Verhaak, R. G. W., McKenna, A., Campos, B., Noushmehr, H., Salama, S. R., Zheng, S., Chakravarty, D., Sanborn, J. Z., Berman, S. H., et al. (2013) · Cell · doi:10.1016/j.cell.2013.09.034 · Cited by 5,016
Deep Learning: Methods and Applications
Deng, L., & Yu, D. (2014) · Foundations and Trends® in Signal Processing · doi:10.1561/2000000039 · Cited by 3,480 · cited 8×
“Although it may be clearer and more explicit to refer to the strategy used in this study as a stacked restricted Boltzmann machines – deep autoencoder (SRBM – DA), we will use the more traditional DBN terminology [1, 13 – 15] for the sake of being consistent with the literature.”
BackgroundMethodInfluentialA Practical Guide to Training Restricted Boltzmann Machines
Hinton, G. E. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8_32 · Cited by 2,852 · cited 1×
“Our strategies for deep learning model selection were guided by articles from Bengio [20] and Hinton [21].”
MethodIdentification of a CpG Island Methylator Phenotype that Defines a Distinct Subgroup of Glioma
Noushmehr, H., Weisenberger, D. J., Diefes, K., Phillips, H. S., Pujara, K., Berman, B. P., Pan, F., Pelloski, C. E., Sulman, E. P., Bhat, K. P., et al. (2010) · Cancer Cell · doi:10.1016/j.ccr.2010.03.017 · Cited by 2,337
Practical Recommendations for Gradient-Based Training of Deep Architectures
Bengio, Y. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8_26 · Cited by 2,259 · cited 2×
“A combined random and grid search approach [20] were used with a goal of finding the set of hyperpara-meters that minimized the average test set reconstruction error and prevented overfitting, while also significantly reducing the dimensionality of the data (i.e., final (top) hidden layer with…”
MethodA census of human transcription factors: function, expression and evolution
Vaquerizas, J. M., Kummerfeld, S. K., Teichmann, S. A., & Luscombe, N. M. (2009) · Nature Reviews Genetics · doi:10.1038/nrg2538 · Cited by 1,573 · cited 1×
“Interestingly, our hypothesis agrees surprisingly well with the current consensus on the number of human TFs (~1400) estimated through analyzing the human genome [25].”
BackgroundIntegrative data analysis of multi-platform cancer data with a multimodal deep learning approach
Liang, M., Li, Z., Chen, T., & Zeng, J. (2015) · IEEE/ACM Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2014.2377729 · Cited by 231
Cited by
Artificial intelligence in cancer imaging: Clinical challenges and applications
Bi, W. L., Hosny, A., Schabath, M. B., Giger, M. L., Birkbak, N. J., Mehrtash, A., Allison, T., Arnaout, O., Abbosh, C., Dunn, I. F., et al. (2019) · CA: A Cancer Journal for Clinicians · doi:10.3322/caac.21552 · Cited by 1,846
Artificial intelligence in drug development: Present status and future prospects
Mak, K.-K., & Pichika, M. R. (2019) · Drug Discovery Today · doi:10.1016/j.drudis.2018.11.014 · Cited by 1,018
A Structure-Based Drug Discovery Paradigm
Batool, M., Ahmad, B., & Choi, S. (2019) · International Journal of Molecular Sciences · doi:10.3390/ijms20112783 · Cited by 744
Deep learning in drug discovery: opportunities, challenges and future prospects
Lavecchia, A. (2019) · Drug Discovery Today · doi:10.1016/j.drudis.2019.07.006 · Cited by 353
Metadata
- Authors
- Young, J. D., Cai, C., & Lu, X.
- Venue
- BMC Bioinformatics
- Published
- 2017-10
- Volume
- 18
- Issue
- S11
- Pages
- 381-381
- ISSN
- 1471-2105 (ISSN-L)
- OpenAlex
- W2761714660
- Cited by
- 70
- References
- 9
- FWCI
- 2.33
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma. https://app.cochat.ai/papers/10.1186/s12859-017-1798-2
- The paper
- Young, J. D., Cai, C., & Lu, X. (2017). Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma. BMC Bioinformatics. https://doi.org/10.1186/s12859-017-1798-2
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.