Unsupervised machine learning reveals risk stratifying glioblastoma tumor cells
Abstract
A goal of cancer research is to reveal cell subsets linked to continuous clinical outcomes to generate new therapeutic and biomarker hypotheses. We introduce a machine learning algorithm, Risk Assessment Population IDentification (RAPID), that is unsupervised and automated, identifies phenotypically distinct cell populations, and determines whether these populations stratify patient survival. With a pilot mass cytometry dataset of 2 million cells from 28 glioblastomas, RAPID identified tumor cells whose abundance independently and continuously stratified patient survival. Statistical validation within the workflow included repeated runs of stochastic steps and cell subsampling. Biological validation used an orthogonal platform, immunohistochemistry, and a larger cohort of 73 glioblastoma patients to confirm the findings from the pilot cohort. RAPID was also validated to find known risk-stratifying cells and features using published data from blood cancer. Thus, RAPID provides an automated, unsupervised approach for finding statistically and biologically significant cells using cytometry data from patient samples.
Data availability
Annotated flow data files are available at the following link: https://flowrepository.org/id/FR-FCM-Z24K. Patient specific views of population abundance and channel mass signals for all analyzed patients in this study are currently available in Supplementary File 6. RAPID code is currently available on Github, together with example analysis data: https://github.com/cytolab/RAPID
-
Single-cell developmental classification of B cell precursor acute lymphoblastic leukemia at diagnosis reveals predictors of relapseGithub Mass cytometry data for DDPR project.
Article and author information
Author details
Funding
National Institutes of Health (R00 CA143231)
- Jonathan M Irish
National Institutes of Health (F31 CA199993)
- Allison R Greenplate
National Institutes of Health (R25 CA136440-04)
- Kirsten E Diggins
Vanderbilt Ingram Cancer Center (Provocative Question)
- Jonathan M Irish
National Institutes of Health (R01 CA226833)
- Jonathan M Irish
National Institutes of Health (U54 CA217450)
- Jonathan M Irish
National Institutes of Health (U01 AI125056)
- Sierra M Barone
- Jonathan M Irish
National Institutes of Health (R01 NS096238)
- Rebecca A Ihrie
U.S. Department of Defense (W81XWH-16-1-0171)
- Rebecca A Ihrie
Michael David Greene Brain Cancer Fund
- Rebecca A Ihrie
- Jonathan M Irish
Vanderbilt Institute for Clinical and Translational Research (VR51342)
- Bret C Mobley
- Rebecca A Ihrie
Vanderbilt Ingram Cancer Center (P30 CA68485)
- Jonathan M Irish
Vanderbilt Ingram Cancer Center (Ambassadors Award)
- Rebecca A Ihrie
- Jonathan M Irish
Southeastern Brain Tumor Foundation
- Rebecca A Ihrie
- Jonathan M Irish
Vanderbilt University (International Scholars Program)
- Nalin Leelatian
Vanderbilt University (Discovery Grant)
- Nalin Leelatian
- Jonathan M Irish
Alpha Omega Alpha Honor Medical Society (Postgraduate Award)
- Akshitkumar M Mistry
Society of Neurological Surgeons (RUNN Award)
- Akshitkumar M Mistry
National Institutes of Health (F32 CA224962-01)
- Akshitkumar M Mistry
Burroughs Wellcome Fund (1018894)
- Akshitkumar M Mistry
National Institutes of Health (T32 HD007502)
- Justine Sinnaeve
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Reviewing Editor
- C Daniela Robles-Espinoza, International Laboratory for Human Genome Research, Mexico
Version history
- Received: March 12, 2020
- Accepted: June 4, 2020
- Accepted Manuscript published: June 23, 2020 (version 1)
- Version of Record published: July 7, 2020 (version 2)
Copyright
© 2020, Leelatian et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 3,576
- Page views
-
- 443
- Downloads
-
- 15
- Citations
Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.
Download links
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Further reading
-
- Cell Biology
- Computational and Systems Biology
Computer models of the human ventricular cardiomyocyte action potential (AP) have reached a level of detail and maturity that has led to an increasing number of applications in the pharmaceutical sector. However, interfacing the models with experimental data can become a significant computational burden. To mitigate the computational burden, the present study introduces a neural network (NN) that emulates the AP for given maximum conductances of selected ion channels, pumps, and exchangers. Its applicability in pharmacological studies was tested on synthetic and experimental data. The NN emulator potentially enables massive speed-ups compared to regular simulations and the forward problem (find drugged AP for pharmacological parameters defined as scaling factors of control maximum conductances) on synthetic data could be solved with average root-mean-square errors (RMSE) of 0.47 mV in normal APs and of 14.5 mV in abnormal APs exhibiting early afterdepolarizations (72.5% of the emulated APs were alining with the abnormality, and the substantial majority of the remaining APs demonstrated pronounced proximity). This demonstrates not only very fast and mostly very accurate AP emulations but also the capability of accounting for discontinuities, a major advantage over existing emulation strategies. Furthermore, the inverse problem (find pharmacological parameters for control and drugged APs through optimization) on synthetic data could be solved with high accuracy shown by a maximum RMSE of 0.22 in the estimated pharmacological parameters. However, notable mismatches were observed between pharmacological parameters estimated from experimental data and distributions obtained from the Comprehensive in vitro Proarrhythmia Assay initiative. This reveals larger inaccuracies which can be attributed particularly to the fact that small tissue preparations were studied while the emulator was trained on single cardiomyocyte data. Overall, our study highlights the potential of NN emulators as powerful tool for an increased efficiency in future quantitative systems pharmacology studies.
-
- Computational and Systems Biology
- Neuroscience
Closed-loop neuronal stimulation has a strong therapeutic potential for neurological disorders such as Parkinson’s disease. However, at the moment, standard stimulation protocols rely on continuous open-loop stimulation and the design of adaptive controllers is an active field of research. Delayed feedback control (DFC), a popular method used to control chaotic systems, has been proposed as a closed-loop technique for desynchronisation of neuronal populations but, so far, was only tested in computational studies. We implement DFC for the first time in neuronal populations and access its efficacy in disrupting unwanted neuronal oscillations. To analyse in detail the performance of this activity control algorithm, we used specialised in vitro platforms with high spatiotemporal monitoring/stimulating capabilities. We show that the conventional DFC in fact worsens the neuronal population oscillatory behaviour, which was never reported before. Conversely, we present an improved control algorithm, adaptive DFC (aDFC), which monitors the ongoing oscillation periodicity and self-tunes accordingly. aDFC effectively disrupts collective neuronal oscillations restoring a more physiological state. Overall, these results support aDFC as a better candidate for therapeutic closed-loop brain stimulation.