TY - JOUR
T1 - Genetic classification of populations using supervised learning
AU - Bridges, Michael
AU - Heron, Elizabeth A.
AU - O'Dushlaine, Colm
AU - Segurado, Ricardo
AU - Morris, Derek
AU - Corvin, Aiden
AU - Gill, Michael
AU - Pinto, Carlos
AU - Morris, Derek W.
AU - O'Dushlaine, Colm
AU - Kenny, Elaine
AU - Quinn, Emma M.
AU - Gill, Michael
AU - Corvin, Aiden
AU - O'Donovan, Michael C.
AU - Kirov, George K.
AU - Craddock, Nick J.
AU - Holmans, Peter A.
AU - Williams, Nigel M.
AU - Georgieva, Lucy
AU - Nikolov, Ivan
AU - Norton, N.
AU - Williams, H.
AU - Toncheva, Draga
AU - Milanova, Vihra
AU - Owen, Michael J.
AU - Hultman, Christina M.
AU - Lichtenstein, Paul
AU - Thelander, Emma F.
AU - Sullivan, Patrick
AU - McQuillin, Andrew
AU - Choudhury, Khalid
AU - Datta, Susmita
AU - Pimm, Jonathan
AU - Thirumalai, Srinivasa
AU - Puri, Vinay
AU - Krasucki, Robert
AU - Lawrence, Jacob
AU - Quested, Digby
AU - Bass, Nicholas
AU - Gurling, Hugh
AU - Crombie, Caroline
AU - Fraser, Gillian
AU - Kuan, Soh Leh
AU - Walker, Nicholas
AU - St Clair, David
AU - Blackwood, Douglas H.R.
AU - Muir, Walter J.
AU - McGhee, Kevin A.
AU - Pickard, Ben
AU - The International Schizophrenia Consortium (ISC)
N1 - Funding: This project has not been directly funded by any agency. The authors employed on research contracts are supported by the Wellcome Trust (http://www.wellcome.ac.uk), Science Foundation Ireland (http://www.sfi.ie), and the UK Science and Technology Research Council (http://www.stfc.ac.uk). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
PY - 2011/5/12
Y1 - 2011/5/12
N2 - There are many instances in genetics in which we wish to determine whether two candidate populations are distinguishable on the basis of their genetic structure. Examples include populations which are geographically separated, case-control studies and quality control (when participants in a study have been genotyped at different laboratories). This latter application is of particular importance in the era of large scale genome wide association studies, when collections of individuals genotyped at different locations are being merged to provide increased power. The traditional method for detecting structure within a population is some form of exploratory technique such as principal components analysis. Such methods, which do not utilise our prior knowledge of the membership of the candidate populations. are termed unsupervised. Supervised methods, on the other hand are able to utilise this prior knowledge when it is available. In this paper we demonstrate that in such cases modern supervised approaches are a more appropriate tool for detecting genetic differences between populations. We apply two such methods, (neural networks and support vector machines) to the classification of three populations (two from Scotland and one from Bulgaria). The sensitivity exhibited by both these methods is considerably higher than that attained by principal components analysis and in fact comfortably exceeds a recently conjectured theoretical limit on the sensitivity of unsupervised methods. In particular, our methods can distinguish between the two Scottish populations, where principal components analysis cannot. We suggest, on the basis of our results that a supervised learning approach should be the method of choice when classifying individuals into pre-defined populations, particularly in quality control for large scale genome wide association studies.
AB - There are many instances in genetics in which we wish to determine whether two candidate populations are distinguishable on the basis of their genetic structure. Examples include populations which are geographically separated, case-control studies and quality control (when participants in a study have been genotyped at different laboratories). This latter application is of particular importance in the era of large scale genome wide association studies, when collections of individuals genotyped at different locations are being merged to provide increased power. The traditional method for detecting structure within a population is some form of exploratory technique such as principal components analysis. Such methods, which do not utilise our prior knowledge of the membership of the candidate populations. are termed unsupervised. Supervised methods, on the other hand are able to utilise this prior knowledge when it is available. In this paper we demonstrate that in such cases modern supervised approaches are a more appropriate tool for detecting genetic differences between populations. We apply two such methods, (neural networks and support vector machines) to the classification of three populations (two from Scotland and one from Bulgaria). The sensitivity exhibited by both these methods is considerably higher than that attained by principal components analysis and in fact comfortably exceeds a recently conjectured theoretical limit on the sensitivity of unsupervised methods. In particular, our methods can distinguish between the two Scottish populations, where principal components analysis cannot. We suggest, on the basis of our results that a supervised learning approach should be the method of choice when classifying individuals into pre-defined populations, particularly in quality control for large scale genome wide association studies.
UR - http://www.scopus.com/inward/record.url?scp=79955939428&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0014802
DO - 10.1371/journal.pone.0014802
M3 - Article
C2 - 21589856
AN - SCOPUS:79955939428
SN - 1932-6203
VL - 6
SP - 1
EP - 12
JO - PloS ONE
JF - PloS ONE
IS - 5
M1 - e14802
ER -