Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer

Ricardo De Matos Simoes, Constantine Mitsiades, Kathleen Williamson, Frank Emmert-Streib

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Urothelial cancer (UC) is highly recurrent and can progress from non-invasive (NMIUC) to a more aggressive muscle-invasive (MIUC) subtype that invades the muscle tissue layer of the bladder. We present a proof of principle study that network-based features of gene pairs can be used to improve classifier performance and the functional analysis of urothelial cancer gene expression data. In the first step of our procedure each individual sample of a UC gene expression dataset is inflated by gene pair expression ratios that are defined based on a given network structure. In the second step an elastic net feature selection procedure for network-based signatures is applied to discriminate between NMIUC and MIUC samples. We performed a repeated random subsampling cross validation in three independent datasets. The network signatures were characterized by a functional enrichment analysis and studied for the enrichment of known cancer genes. We observed that the network-based gene signatures from meta collections of proteinprotein interaction (PPI) databases such as CPDB and the PPI databases HPRD and BioGrid improved the classification performance compared to single gene based signatures. The network based signatures that were derived from PPI databases showed a prominent enrichment of cancer genes (e.g., TP53, TRIM27 and HNRNPA2Bl). We provide a novel integrative approach for large-scale gene expression analysis for the identification and development of novel diagnostical targets in bladder cancer. Further, our method allowed to link cancer gene associations to network-based expression signatures that are not observed in gene-based expression signatures.

Original languageEnglish
Title of host publicationProceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1216-1223
Number of pages8
ISBN (Print)9781467367981
DOIs
Publication statusPublished - 16 Dec 2015
EventIEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015 - Washington, United States
Duration: 09 Nov 201512 Nov 2015

Conference

ConferenceIEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015
CountryUnited States
CityWashington
Period09/11/201512/11/2015

Fingerprint

Neoplasm Genes
Gene expression
Muscle
Genes
Gene Expression
Muscles
Databases
Neoplasms
Gene Regulatory Networks
Transcriptome
Urinary Bladder Neoplasms
Functional analysis
Urinary Bladder
Feature extraction
Classifiers
Tissue
Datasets

Keywords

  • data feature space inflation
  • feature selection
  • gene pair expression ratio
  • muscle-invasive
  • non muscleinvasive
  • Urothelial cancer

Cite this

De Matos Simoes, R., Mitsiades, C., Williamson, K., & Emmert-Streib, F. (2015). Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer. In Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015 (pp. 1216-1223). [7359855] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM.2015.7359855
De Matos Simoes, Ricardo ; Mitsiades, Constantine ; Williamson, Kathleen ; Emmert-Streib, Frank. / Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer. Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 1216-1223
@inproceedings{79bbcc5ef53d4a80b9a9ddd424862fb8,
title = "Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer",
abstract = "Urothelial cancer (UC) is highly recurrent and can progress from non-invasive (NMIUC) to a more aggressive muscle-invasive (MIUC) subtype that invades the muscle tissue layer of the bladder. We present a proof of principle study that network-based features of gene pairs can be used to improve classifier performance and the functional analysis of urothelial cancer gene expression data. In the first step of our procedure each individual sample of a UC gene expression dataset is inflated by gene pair expression ratios that are defined based on a given network structure. In the second step an elastic net feature selection procedure for network-based signatures is applied to discriminate between NMIUC and MIUC samples. We performed a repeated random subsampling cross validation in three independent datasets. The network signatures were characterized by a functional enrichment analysis and studied for the enrichment of known cancer genes. We observed that the network-based gene signatures from meta collections of proteinprotein interaction (PPI) databases such as CPDB and the PPI databases HPRD and BioGrid improved the classification performance compared to single gene based signatures. The network based signatures that were derived from PPI databases showed a prominent enrichment of cancer genes (e.g., TP53, TRIM27 and HNRNPA2Bl). We provide a novel integrative approach for large-scale gene expression analysis for the identification and development of novel diagnostical targets in bladder cancer. Further, our method allowed to link cancer gene associations to network-based expression signatures that are not observed in gene-based expression signatures.",
keywords = "data feature space inflation, feature selection, gene pair expression ratio, muscle-invasive, non muscleinvasive, Urothelial cancer",
author = "{De Matos Simoes}, Ricardo and Constantine Mitsiades and Kathleen Williamson and Frank Emmert-Streib",
year = "2015",
month = "12",
day = "16",
doi = "10.1109/BIBM.2015.7359855",
language = "English",
isbn = "9781467367981",
pages = "1216--1223",
booktitle = "Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

De Matos Simoes, R, Mitsiades, C, Williamson, K & Emmert-Streib, F 2015, Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer. in Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015., 7359855, Institute of Electrical and Electronics Engineers Inc., pp. 1216-1223, IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015, Washington, United States, 09/11/2015. https://doi.org/10.1109/BIBM.2015.7359855

Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer. / De Matos Simoes, Ricardo; Mitsiades, Constantine; Williamson, Kathleen; Emmert-Streib, Frank.

Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 1216-1223 7359855.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer

AU - De Matos Simoes, Ricardo

AU - Mitsiades, Constantine

AU - Williamson, Kathleen

AU - Emmert-Streib, Frank

PY - 2015/12/16

Y1 - 2015/12/16

N2 - Urothelial cancer (UC) is highly recurrent and can progress from non-invasive (NMIUC) to a more aggressive muscle-invasive (MIUC) subtype that invades the muscle tissue layer of the bladder. We present a proof of principle study that network-based features of gene pairs can be used to improve classifier performance and the functional analysis of urothelial cancer gene expression data. In the first step of our procedure each individual sample of a UC gene expression dataset is inflated by gene pair expression ratios that are defined based on a given network structure. In the second step an elastic net feature selection procedure for network-based signatures is applied to discriminate between NMIUC and MIUC samples. We performed a repeated random subsampling cross validation in three independent datasets. The network signatures were characterized by a functional enrichment analysis and studied for the enrichment of known cancer genes. We observed that the network-based gene signatures from meta collections of proteinprotein interaction (PPI) databases such as CPDB and the PPI databases HPRD and BioGrid improved the classification performance compared to single gene based signatures. The network based signatures that were derived from PPI databases showed a prominent enrichment of cancer genes (e.g., TP53, TRIM27 and HNRNPA2Bl). We provide a novel integrative approach for large-scale gene expression analysis for the identification and development of novel diagnostical targets in bladder cancer. Further, our method allowed to link cancer gene associations to network-based expression signatures that are not observed in gene-based expression signatures.

AB - Urothelial cancer (UC) is highly recurrent and can progress from non-invasive (NMIUC) to a more aggressive muscle-invasive (MIUC) subtype that invades the muscle tissue layer of the bladder. We present a proof of principle study that network-based features of gene pairs can be used to improve classifier performance and the functional analysis of urothelial cancer gene expression data. In the first step of our procedure each individual sample of a UC gene expression dataset is inflated by gene pair expression ratios that are defined based on a given network structure. In the second step an elastic net feature selection procedure for network-based signatures is applied to discriminate between NMIUC and MIUC samples. We performed a repeated random subsampling cross validation in three independent datasets. The network signatures were characterized by a functional enrichment analysis and studied for the enrichment of known cancer genes. We observed that the network-based gene signatures from meta collections of proteinprotein interaction (PPI) databases such as CPDB and the PPI databases HPRD and BioGrid improved the classification performance compared to single gene based signatures. The network based signatures that were derived from PPI databases showed a prominent enrichment of cancer genes (e.g., TP53, TRIM27 and HNRNPA2Bl). We provide a novel integrative approach for large-scale gene expression analysis for the identification and development of novel diagnostical targets in bladder cancer. Further, our method allowed to link cancer gene associations to network-based expression signatures that are not observed in gene-based expression signatures.

KW - data feature space inflation

KW - feature selection

KW - gene pair expression ratio

KW - muscle-invasive

KW - non muscleinvasive

KW - Urothelial cancer

UR - http://www.scopus.com/inward/record.url?scp=84962439353&partnerID=8YFLogxK

U2 - 10.1109/BIBM.2015.7359855

DO - 10.1109/BIBM.2015.7359855

M3 - Conference contribution

AN - SCOPUS:84962439353

SN - 9781467367981

SP - 1216

EP - 1223

BT - Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

De Matos Simoes R, Mitsiades C, Williamson K, Emmert-Streib F. Network signatures based on gene pair expression ratios improve classification and the analysis of muscle-invasive urothelial cancer. In Proceedings - 2015 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 1216-1223. 7359855 https://doi.org/10.1109/BIBM.2015.7359855