A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State
Gliomas are the most common primary neoplasm of the central nervous system. A promising frontier in the definition of glioma prognosis and treatment is represented by epigenetics. Furthermore, in this study, we developed a machine learning classification model based on epigenetic data (CpG probes) t...
Main Authors: | , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Cells |
Subjects: | |
Online Access: | https://www.mdpi.com/2073-4409/10/3/576 |
id |
doaj-0ae2d1106c164a4b8a8c848a6b78b1db |
---|---|
record_format |
Article |
spelling |
doaj-0ae2d1106c164a4b8a8c848a6b78b1db2021-03-06T00:09:22ZengMDPI AGCells2073-44092021-03-011057657610.3390/cells10030576A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive StateMaurizio Polano0Emanuele Fabbiani1Eva Adreuzzi2Federica Di Cintio3Luca Bedon4Davide Gentilini5Maurizio Mongiat6Tamara Ius7Mauro Arcicasa8Miran Skrap9Michele Dal Bo10Giuseppe Toffoli11Experimental and Clinical Pharmacology Unit, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyDepartment of Electrical, Computer and Biomedical Engineering,University of Pavia, 27100 Pavia, ItalyDivision of Molecular Oncology, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyExperimental and Clinical Pharmacology Unit, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyExperimental and Clinical Pharmacology Unit, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyIstituto Auxologico Italiano IRCCS, Bioinformatics and Statistical Genomics Unit, 20095 Cusano Milanino, ItalyDivision of Molecular Oncology, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyNeurosurgery Unit, Department of Neuroscience, Santa Maria della Misericordia University Hospital, 33100 Udine, ItalyCentro di Riferimento Oncologico di Aviano (CRO), Department of Radiotherapy, IRCCS, 33081 Aviano, ItalyNeurosurgery Unit, Department of Neuroscience, Santa Maria della Misericordia University Hospital, 33100 Udine, ItalyExperimental and Clinical Pharmacology Unit, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyExperimental and Clinical Pharmacology Unit, Centro di Riferimento Oncologico di Aviano (CRO) IRCCS, 33081 Aviano, ItalyGliomas are the most common primary neoplasm of the central nervous system. A promising frontier in the definition of glioma prognosis and treatment is represented by epigenetics. Furthermore, in this study, we developed a machine learning classification model based on epigenetic data (CpG probes) to separate patients according to their state of immunosuppression. We considered 573 cases of low-grade glioma (LGG) and glioblastoma (GBM) from The Cancer Genome Atlas (TCGA). First, from gene expression data, we derived a novel binary indicator to flag patients with a favorable immune state. Then, based on previous studies, we selected the genes related to the immune state of tumor microenvironment. After, we improved the selection with a data-driven procedure, based on Boruta. Finally, we tuned, trained, and evaluated both random forest and neural network classifiers on the resulting dataset. We found that a multi-layer perceptron network fed by the 338 probes selected by applying both expert choice and Boruta results in the best performance, achieving an out-of-sample accuracy of 82.8%, a Matthews correlation coefficient of 0.657, and an area under the ROC curve of 0.9. Based on the proposed model, we provided a method to stratify glioma patients according to their epigenomic state.https://www.mdpi.com/2073-4409/10/3/576immunosuppressiontumor microenviromentneural networkgenome-wide methylation modelgliomaextracellular matrix |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Maurizio Polano Emanuele Fabbiani Eva Adreuzzi Federica Di Cintio Luca Bedon Davide Gentilini Maurizio Mongiat Tamara Ius Mauro Arcicasa Miran Skrap Michele Dal Bo Giuseppe Toffoli |
spellingShingle |
Maurizio Polano Emanuele Fabbiani Eva Adreuzzi Federica Di Cintio Luca Bedon Davide Gentilini Maurizio Mongiat Tamara Ius Mauro Arcicasa Miran Skrap Michele Dal Bo Giuseppe Toffoli A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State Cells immunosuppression tumor microenviroment neural network genome-wide methylation model glioma extracellular matrix |
author_facet |
Maurizio Polano Emanuele Fabbiani Eva Adreuzzi Federica Di Cintio Luca Bedon Davide Gentilini Maurizio Mongiat Tamara Ius Mauro Arcicasa Miran Skrap Michele Dal Bo Giuseppe Toffoli |
author_sort |
Maurizio Polano |
title |
A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State |
title_short |
A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State |
title_full |
A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State |
title_fullStr |
A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State |
title_full_unstemmed |
A New Epigenetic Model to Stratify Glioma Patients According to Their Immunosuppressive State |
title_sort |
new epigenetic model to stratify glioma patients according to their immunosuppressive state |
publisher |
MDPI AG |
series |
Cells |
issn |
2073-4409 |
publishDate |
2021-03-01 |
description |
Gliomas are the most common primary neoplasm of the central nervous system. A promising frontier in the definition of glioma prognosis and treatment is represented by epigenetics. Furthermore, in this study, we developed a machine learning classification model based on epigenetic data (CpG probes) to separate patients according to their state of immunosuppression. We considered 573 cases of low-grade glioma (LGG) and glioblastoma (GBM) from The Cancer Genome Atlas (TCGA). First, from gene expression data, we derived a novel binary indicator to flag patients with a favorable immune state. Then, based on previous studies, we selected the genes related to the immune state of tumor microenvironment. After, we improved the selection with a data-driven procedure, based on Boruta. Finally, we tuned, trained, and evaluated both random forest and neural network classifiers on the resulting dataset. We found that a multi-layer perceptron network fed by the 338 probes selected by applying both expert choice and Boruta results in the best performance, achieving an out-of-sample accuracy of 82.8%, a Matthews correlation coefficient of 0.657, and an area under the ROC curve of 0.9. Based on the proposed model, we provided a method to stratify glioma patients according to their epigenomic state. |
topic |
immunosuppression tumor microenviroment neural network genome-wide methylation model glioma extracellular matrix |
url |
https://www.mdpi.com/2073-4409/10/3/576 |
work_keys_str_mv |
AT mauriziopolano anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT emanuelefabbiani anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT evaadreuzzi anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT federicadicintio anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT lucabedon anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT davidegentilini anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT mauriziomongiat anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT tamaraius anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT mauroarcicasa anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT miranskrap anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT micheledalbo anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT giuseppetoffoli anewepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT mauriziopolano newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT emanuelefabbiani newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT evaadreuzzi newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT federicadicintio newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT lucabedon newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT davidegentilini newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT mauriziomongiat newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT tamaraius newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT mauroarcicasa newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT miranskrap newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT micheledalbo newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate AT giuseppetoffoli newepigeneticmodeltostratifygliomapatientsaccordingtotheirimmunosuppressivestate |
_version_ |
1724229810942640128 |