Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction

Zeliha Gormez; Olcay Kursun; Ahmet Sertbas; Nizamettin Aydin; Huseyin Seker

doi:10.1109/BHI.2012.6211658

Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction

Zeliha Gormez^*, Olcay Kursun, Ahmet Sertbas, Nizamettin Aydin, Huseyin Seker

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

In exploratory association studies of genes with certain diseases, a single or a small number of genes (features) related with the diseases are selected ¹ among many thousands investigated. We investigate the statistical bias and variance of simple yet common (correlation and mutual information based) feature selection algorithms using well-known cross-validation methods (leave-one-out and k-fold) on a gene finding study for hypertension prediction. Our findings show that selected genes are different for different methods and different cross-validation runs for both single gene selection and gene subset selection.

Original language	English
Title of host publication	Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics
Subtitle of host publication	Global Grand Challenge of Health Informatics, BHI 2012
Pages	616-619
Number of pages	4
DOIs	https://doi.org/10.1109/BHI.2012.6211658
Publication status	Published - 2012
Externally published	Yes
Event	IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering - Hong Kong and Shenzhen, China Duration: 2 Jan 2012 → 7 Jan 2012

Publication series

Name	Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012

Conference

Conference	IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering
Country/Territory	China
City	Hong Kong and Shenzhen
Period	2/01/12 → 7/01/12

Access to Document

10.1109/BHI.2012.6211658

Cite this

Gormez, Z., Kursun, O., Sertbas, A., Aydin, N., & Seker, H. (2012). Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction. In Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012 (pp. 616-619). Article 6211658 (Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012). https://doi.org/10.1109/BHI.2012.6211658

Gormez, Zeliha ; Kursun, Olcay ; Sertbas, Ahmet et al. / Statistical bias and variance of gene selection and cross validation methods : A case study on hypertension prediction. Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. pp. 616-619 (Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012).

@inproceedings{75f0d782d86b4b0cbc487836cf228252,

title = "Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction",

abstract = "In exploratory association studies of genes with certain diseases, a single or a small number of genes (features) related with the diseases are selected 1 among many thousands investigated. We investigate the statistical bias and variance of simple yet common (correlation and mutual information based) feature selection algorithms using well-known cross-validation methods (leave-one-out and k-fold) on a gene finding study for hypertension prediction. Our findings show that selected genes are different for different methods and different cross-validation runs for both single gene selection and gene subset selection.",

author = "Zeliha Gormez and Olcay Kursun and Ahmet Sertbas and Nizamettin Aydin and Huseyin Seker",

year = "2012",

doi = "10.1109/BHI.2012.6211658",

language = "English",

isbn = "9781457721779",

series = "Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012",

pages = "616--619",

booktitle = "Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics",

note = "IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering ; Conference date: 02-01-2012 Through 07-01-2012",

}

Gormez, Z, Kursun, O, Sertbas, A, Aydin, N & Seker, H 2012, Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction. in Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012., 6211658, Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012, pp. 616-619, IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering, Hong Kong and Shenzhen, China, 2/01/12. https://doi.org/10.1109/BHI.2012.6211658

Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction. / Gormez, Zeliha; Kursun, Olcay; Sertbas, Ahmet et al.
Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. p. 616-619 6211658 (Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Statistical bias and variance of gene selection and cross validation methods

T2 - IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering

AU - Gormez, Zeliha

AU - Kursun, Olcay

AU - Sertbas, Ahmet

AU - Aydin, Nizamettin

AU - Seker, Huseyin

PY - 2012

Y1 - 2012

N2 - In exploratory association studies of genes with certain diseases, a single or a small number of genes (features) related with the diseases are selected 1 among many thousands investigated. We investigate the statistical bias and variance of simple yet common (correlation and mutual information based) feature selection algorithms using well-known cross-validation methods (leave-one-out and k-fold) on a gene finding study for hypertension prediction. Our findings show that selected genes are different for different methods and different cross-validation runs for both single gene selection and gene subset selection.

AB - In exploratory association studies of genes with certain diseases, a single or a small number of genes (features) related with the diseases are selected 1 among many thousands investigated. We investigate the statistical bias and variance of simple yet common (correlation and mutual information based) feature selection algorithms using well-known cross-validation methods (leave-one-out and k-fold) on a gene finding study for hypertension prediction. Our findings show that selected genes are different for different methods and different cross-validation runs for both single gene selection and gene subset selection.

UR - http://www.scopus.com/inward/record.url?scp=84864264369&partnerID=8YFLogxK

U2 - 10.1109/BHI.2012.6211658

DO - 10.1109/BHI.2012.6211658

M3 - Conference contribution

AN - SCOPUS:84864264369

SN - 9781457721779

T3 - Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012

SP - 616

EP - 619

BT - Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics

Y2 - 2 January 2012 through 7 January 2012

ER -

Gormez Z, Kursun O, Sertbas A, Aydin N, Seker H. Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction. In Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. p. 616-619. 6211658. (Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012). doi: 10.1109/BHI.2012.6211658

Statistical bias and variance of gene selection and cross validation methods: A case study on hypertension prediction

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this