Application of high-dimensional feature selection: evaluation for genomic prediction in man

Mairead Lesley Bermingham; Ricardo Pong-Wong; Athina Spiliopoulou; Caroline Hayward; Igor Rudan; Harry Campbell; Alan F. Wright; Jim Wilson; Felix Agakov; Pau Navarro; Chris Haley

doi:10.1038/srep10312

Application of high-dimensional feature selection: evaluation for genomic prediction in man

Mairead Lesley Bermingham, Ricardo Pong-Wong, Athina Spiliopoulou, Caroline Hayward, Igor Rudan, Harry Campbell, Alan F. Wright, Jim Wilson, Felix Agakov, Pau Navarro, Chris Haley

Research output: Contribution to journal › Article › peer-review

Abstract / Description of output

In this study, we investigated the effect of five feature selection approaches on the performance of a mixed model (G-BLUP) and a Bayesian (Bayes C) prediction method. We predicted height, high density lipoprotein cholesterol (HDL) and body mass index (BMI) within 2,186 Croatian and into 810 UK individuals using genome-wide SNP data. Using all SNP information Bayes C and G-BLUP had similar predictive performance across all traits within the Croatian data, and for the highly polygenic traits height and BMI when predicting into the UK data. Bayes C outperformed G-BLUP in the prediction of HDL, which is influenced by loci of moderate size, in the UK data. Supervised feature selection of a SNP subset in the G-BLUP framework provided a flexible, generalisable and computationally efficient alternative to Bayes C; but careful evaluation of predictive performance is required when supervised feature selection has been used.

Original language	English
Article number	10312
Pages (from-to)	1-12
Number of pages	12
Journal	Scientific Reports
Volume	5
DOIs	https://doi.org/10.1038/srep10312
Publication status	Published - 19 May 2015

Access to Document

10.1038/srep10312

Application of high-dimensional feature selection - publisher's pdf
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
Final published version, 807 KBLicence: Creative Commons: Attribution (CC-BY)

http://www.nature.com/srep/2015/150519/srep10312/full/srep10312.html

A toolbox for the promotion of healthy ageing: Phenotypic prediction from genes and
Haley, C., Agakov, F., Tenesa, A., Woolliams, J., Bermingham, M., Navarro, P., Pong-Wong, R. & Spiliopoulou, A.
BBSRC
1/04/12 → 31/03/15
Project: Research
Orkney Complex Disease Study ORCADES Cardiovascular Study
Wilson, J., Campbell, H. & Wild, S.
UK central government bodies/local authorities, health and hospital authorities
1/04/09 → 31/12/12
Project: Research
Orkney Cardiovascular Disease Study
Wilson, J., Campbell, H. & Webb, D.
UK central government bodies/local authorities, health and hospital authorities
1/10/04 → 30/09/07
Project: Research

Ricardo Pong-Wong
- ricardo.pong-wongroslin.ed.acuk
- Royal (Dick) School of Veterinary Studies - Core Scientist in quantitative predictive biology
Person: Academic: Research Active

Cite this

@article{b907760984394cb886571c7f8a273456,

title = "Application of high-dimensional feature selection: evaluation for genomic prediction in man",

abstract = "In this study, we investigated the effect of five feature selection approaches on the performance of a mixed model (G-BLUP) and a Bayesian (Bayes C) prediction method. We predicted height, high density lipoprotein cholesterol (HDL) and body mass index (BMI) within 2,186 Croatian and into 810 UK individuals using genome-wide SNP data. Using all SNP information Bayes C and G-BLUP had similar predictive performance across all traits within the Croatian data, and for the highly polygenic traits height and BMI when predicting into the UK data. Bayes C outperformed G-BLUP in the prediction of HDL, which is influenced by loci of moderate size, in the UK data. Supervised feature selection of a SNP subset in the G-BLUP framework provided a flexible, generalisable and computationally efficient alternative to Bayes C; but careful evaluation of predictive performance is required when supervised feature selection has been used.",

author = "Bermingham, {Mairead Lesley} and Ricardo Pong-Wong and Athina Spiliopoulou and Caroline Hayward and Igor Rudan and Harry Campbell and Wright, {Alan F.} and Jim Wilson and Felix Agakov and Pau Navarro and Chris Haley",

year = "2015",

month = may,

day = "19",

doi = "10.1038/srep10312",

language = "English",

volume = "5",

pages = "1--12",

journal = "Scientific Reports",

issn = "2045-2322",

publisher = "Nature Publishing Group",

}

TY - JOUR

T1 - Application of high-dimensional feature selection: evaluation for genomic prediction in man

AU - Bermingham, Mairead Lesley

AU - Pong-Wong, Ricardo

AU - Spiliopoulou, Athina

AU - Hayward, Caroline

AU - Rudan, Igor

AU - Campbell, Harry

AU - Wright, Alan F.

AU - Wilson, Jim

AU - Agakov, Felix

AU - Navarro, Pau

AU - Haley, Chris

PY - 2015/5/19

Y1 - 2015/5/19

N2 - In this study, we investigated the effect of five feature selection approaches on the performance of a mixed model (G-BLUP) and a Bayesian (Bayes C) prediction method. We predicted height, high density lipoprotein cholesterol (HDL) and body mass index (BMI) within 2,186 Croatian and into 810 UK individuals using genome-wide SNP data. Using all SNP information Bayes C and G-BLUP had similar predictive performance across all traits within the Croatian data, and for the highly polygenic traits height and BMI when predicting into the UK data. Bayes C outperformed G-BLUP in the prediction of HDL, which is influenced by loci of moderate size, in the UK data. Supervised feature selection of a SNP subset in the G-BLUP framework provided a flexible, generalisable and computationally efficient alternative to Bayes C; but careful evaluation of predictive performance is required when supervised feature selection has been used.

AB - In this study, we investigated the effect of five feature selection approaches on the performance of a mixed model (G-BLUP) and a Bayesian (Bayes C) prediction method. We predicted height, high density lipoprotein cholesterol (HDL) and body mass index (BMI) within 2,186 Croatian and into 810 UK individuals using genome-wide SNP data. Using all SNP information Bayes C and G-BLUP had similar predictive performance across all traits within the Croatian data, and for the highly polygenic traits height and BMI when predicting into the UK data. Bayes C outperformed G-BLUP in the prediction of HDL, which is influenced by loci of moderate size, in the UK data. Supervised feature selection of a SNP subset in the G-BLUP framework provided a flexible, generalisable and computationally efficient alternative to Bayes C; but careful evaluation of predictive performance is required when supervised feature selection has been used.

U2 - 10.1038/srep10312

DO - 10.1038/srep10312

M3 - Article

SN - 2045-2322

VL - 5

SP - 1

EP - 12

JO - Scientific Reports

JF - Scientific Reports

M1 - 10312

ER -

Application of high-dimensional feature selection: evaluation for genomic prediction in man

Abstract / Description of output

Access to Document

Fingerprint

Projects

A toolbox for the promotion of healthy ageing: Phenotypic prediction from genes and

Orkney Complex Disease Study ORCADES Cardiovascular Study

Orkney Cardiovascular Disease Study

Profiles

Ricardo Pong-Wong

Cite this