MiRNA and gene expression based cancer classification using self-learning and co-training approaches

Rania Ibrahim, Noha Yousri, Mohamed A. Ismail, Nagwa M. El-Makky

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

8 Citations (Scopus)

Abstract

A number of attempts to classify cancer samples using miRNA/gene expression profiles are known in literature. However, semi-supervised learning models have only been recently introduced to exploit the huge unlabeled expression profiles in enhancing sample classification. It is important to combine both miRNA and gene expression sets as that provides more information on the characteristics of cancer samples. The use of both of labeled and unlabeled miRNA and gene expression sets to enhance sample classification has not been explored yet. In this paper, two semi-supervised machine learning approaches, namely self-learning and co-training are adapted to enhance the quality of cancer sample classification. In self-learning, miRNA and gene based classifiers are enhanced independently. While in co-training, both miRNA and gene expression profiles are used simultaneously to provide different views of cancer samples. The approaches were evaluated using breast cancer, hepatocellular carcinoma (HCC) and lung cancer expression sets. Results show up to 20% improvement in F1-measure over Random Forests and SVM classifiers. Co-Training also outperforms Low Density Separation (LDS) approach by around 25% improvement in F1-measure in breast cancer.

Original languageEnglish
Title of host publicationProceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013
Pages495-498
Number of pages4
DOIs
Publication statusPublished - 1 Dec 2013
Externally publishedYes
Event2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013 - Shanghai, China
Duration: 18 Dec 201321 Dec 2013

Other

Other2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013
CountryChina
CityShanghai
Period18/12/1321/12/13

Fingerprint

Gene expression
Classifiers
Supervised learning
Learning systems
Genes

Keywords

  • Cancer sample classifiers
  • Co-Training
  • miRNA and gene expression analysis
  • Self-Learning
  • Semi-supervised Approaches

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Ibrahim, R., Yousri, N., Ismail, M. A., & El-Makky, N. M. (2013). MiRNA and gene expression based cancer classification using self-learning and co-training approaches. In Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013 (pp. 495-498). [6732544] https://doi.org/10.1109/BIBM.2013.6732544

MiRNA and gene expression based cancer classification using self-learning and co-training approaches. / Ibrahim, Rania; Yousri, Noha; Ismail, Mohamed A.; El-Makky, Nagwa M.

Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. p. 495-498 6732544.

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

Ibrahim, R, Yousri, N, Ismail, MA & El-Makky, NM 2013, MiRNA and gene expression based cancer classification using self-learning and co-training approaches. in Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013., 6732544, pp. 495-498, 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013, Shanghai, China, 18/12/13. https://doi.org/10.1109/BIBM.2013.6732544
Ibrahim R, Yousri N, Ismail MA, El-Makky NM. MiRNA and gene expression based cancer classification using self-learning and co-training approaches. In Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. p. 495-498. 6732544 https://doi.org/10.1109/BIBM.2013.6732544
Ibrahim, Rania ; Yousri, Noha ; Ismail, Mohamed A. ; El-Makky, Nagwa M. / MiRNA and gene expression based cancer classification using self-learning and co-training approaches. Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. pp. 495-498
@inproceedings{5006440366594491b30fd55307f3ea13,
title = "MiRNA and gene expression based cancer classification using self-learning and co-training approaches",
abstract = "A number of attempts to classify cancer samples using miRNA/gene expression profiles are known in literature. However, semi-supervised learning models have only been recently introduced to exploit the huge unlabeled expression profiles in enhancing sample classification. It is important to combine both miRNA and gene expression sets as that provides more information on the characteristics of cancer samples. The use of both of labeled and unlabeled miRNA and gene expression sets to enhance sample classification has not been explored yet. In this paper, two semi-supervised machine learning approaches, namely self-learning and co-training are adapted to enhance the quality of cancer sample classification. In self-learning, miRNA and gene based classifiers are enhanced independently. While in co-training, both miRNA and gene expression profiles are used simultaneously to provide different views of cancer samples. The approaches were evaluated using breast cancer, hepatocellular carcinoma (HCC) and lung cancer expression sets. Results show up to 20{\%} improvement in F1-measure over Random Forests and SVM classifiers. Co-Training also outperforms Low Density Separation (LDS) approach by around 25{\%} improvement in F1-measure in breast cancer.",
keywords = "Cancer sample classifiers, Co-Training, miRNA and gene expression analysis, Self-Learning, Semi-supervised Approaches",
author = "Rania Ibrahim and Noha Yousri and Ismail, {Mohamed A.} and El-Makky, {Nagwa M.}",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/BIBM.2013.6732544",
language = "English",
isbn = "9781479913091",
pages = "495--498",
booktitle = "Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013",

}

TY - GEN

T1 - MiRNA and gene expression based cancer classification using self-learning and co-training approaches

AU - Ibrahim, Rania

AU - Yousri, Noha

AU - Ismail, Mohamed A.

AU - El-Makky, Nagwa M.

PY - 2013/12/1

Y1 - 2013/12/1

N2 - A number of attempts to classify cancer samples using miRNA/gene expression profiles are known in literature. However, semi-supervised learning models have only been recently introduced to exploit the huge unlabeled expression profiles in enhancing sample classification. It is important to combine both miRNA and gene expression sets as that provides more information on the characteristics of cancer samples. The use of both of labeled and unlabeled miRNA and gene expression sets to enhance sample classification has not been explored yet. In this paper, two semi-supervised machine learning approaches, namely self-learning and co-training are adapted to enhance the quality of cancer sample classification. In self-learning, miRNA and gene based classifiers are enhanced independently. While in co-training, both miRNA and gene expression profiles are used simultaneously to provide different views of cancer samples. The approaches were evaluated using breast cancer, hepatocellular carcinoma (HCC) and lung cancer expression sets. Results show up to 20% improvement in F1-measure over Random Forests and SVM classifiers. Co-Training also outperforms Low Density Separation (LDS) approach by around 25% improvement in F1-measure in breast cancer.

AB - A number of attempts to classify cancer samples using miRNA/gene expression profiles are known in literature. However, semi-supervised learning models have only been recently introduced to exploit the huge unlabeled expression profiles in enhancing sample classification. It is important to combine both miRNA and gene expression sets as that provides more information on the characteristics of cancer samples. The use of both of labeled and unlabeled miRNA and gene expression sets to enhance sample classification has not been explored yet. In this paper, two semi-supervised machine learning approaches, namely self-learning and co-training are adapted to enhance the quality of cancer sample classification. In self-learning, miRNA and gene based classifiers are enhanced independently. While in co-training, both miRNA and gene expression profiles are used simultaneously to provide different views of cancer samples. The approaches were evaluated using breast cancer, hepatocellular carcinoma (HCC) and lung cancer expression sets. Results show up to 20% improvement in F1-measure over Random Forests and SVM classifiers. Co-Training also outperforms Low Density Separation (LDS) approach by around 25% improvement in F1-measure in breast cancer.

KW - Cancer sample classifiers

KW - Co-Training

KW - miRNA and gene expression analysis

KW - Self-Learning

KW - Semi-supervised Approaches

UR - http://www.scopus.com/inward/record.url?scp=84894520755&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894520755&partnerID=8YFLogxK

U2 - 10.1109/BIBM.2013.6732544

DO - 10.1109/BIBM.2013.6732544

M3 - Conference contribution

AN - SCOPUS:84894520755

SN - 9781479913091

SP - 495

EP - 498

BT - Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013

ER -