Application of Two Machine Learning Algorithms to Genetic Association Studies in the Presence of Covariates

Overview

Journal BMC Genet

Publisher Biomed Central

Specialties Biotechnology
Molecular Biology

Date 2008 Nov 19

PMID 19014573

Citations 8

Authors

Bareng A S Nonyane

Andrea S Foulkes

Affiliations

Soon will be listed here.

Abstract

Background: Population-based investigations aimed at uncovering genotype-trait associations often involve high-dimensional genetic polymorphism data as well as information on multiple environmental and clinical parameters. Machine learning (ML) algorithms offer a straightforward analytic approach for selecting subsets of these inputs that are most predictive of a pre-defined trait. The performance of these algorithms, however, in the presence of covariates is not well characterized.

Methods And Results: In this manuscript, we investigate two approaches: Random Forests (RFs) and Multivariate Adaptive Regression Splines (MARS). Through multiple simulation studies, the performance under several underlying models is evaluated. An application to a cohort of HIV-1 infected individuals receiving anti-retroviral therapies is also provided.

Conclusion: Consistent with more traditional regression modeling theory, our findings highlight the importance of considering the nature of underlying gene-covariate-trait relationships before applying ML algorithms, particularly when there is potential confounding or effect mediation.

Citing Articles

Covariate adjusted classification trees.

Asafu-Adjei J, Sampson A Biostatistics. 2017; 19(1):42-53.

PMID: 28520903 PMC: 6075597. DOI: 10.1093/biostatistics/kxx015.

RAPIDSNPs: A new computational pipeline for rapidly identifying key genetic variants reveals previously unidentified SNPs that are significantly associated with individual platelet responses.

Salehe B, Jones C, Di Fatta G, McGuffin L PLoS One. 2017; 12(4):e0175957.

PMID: 28441463 PMC: 5404774. DOI: 10.1371/journal.pone.0175957.

EPAS1 gene variants are associated with sprint/power athletic performance in two cohorts of European athletes.

Voisin S, Cieszczyk P, Pushkarev V, Dyatlov D, Vashlyayev B, Shumaylov V BMC Genomics. 2014; 15:382.

PMID: 24884370 PMC: 4035083. DOI: 10.1186/1471-2164-15-382.

Integrative systems biology approaches in asthma pharmacogenomics.

Dahlin A, Tantisira K Pharmacogenomics. 2012; 13(12):1387-404.

PMID: 22966888 PMC: 3553555. DOI: 10.2217/pgs.12.126.

An integrated approach to reduce the impact of minor allele frequency and linkage disequilibrium on variable importance measures for genome-wide data.

Walters R, Laurin C, Lubke G Bioinformatics. 2012; 28(20):2615-23.

PMID: 22847933 PMC: 3467741. DOI: 10.1093/bioinformatics/bts483.

References

Foulkes A, Wohl D, Frank I, Puleo E, Restine S, Wolfe M . Associations among race/ethnicity, ApoC-III genotypes, and lipids in HIV-1-infected individuals on antiretroviral therapy. PLoS Med. 2006; 3(3):e52. PMC: 1334223. DOI: 10.1371/journal.pmed.0030052. View

Christenfeld N, Sloan R, Carroll D, Greenland S . Risk factors, confounding, and the illusion of statistical control. Psychosom Med. 2004; 66(6):868-75. DOI: 10.1097/01.psy.0000140008.70959.41. View

Tan C, Tai E, Tan C, Chia K, Lee J, Chew S . APOE polymorphism and lipid profile in three ethnic groups in the Singapore population. Atherosclerosis. 2003; 170(2):253-60. DOI: 10.1016/s0021-9150(03)00232-6. View

Lunetta K, Hayward L, Segal J, Van Eerdewegh P . Screening large-scale association study data: exploiting interactions using random forests. BMC Genet. 2004; 5:32. PMC: 545646. DOI: 10.1186/1471-2156-5-32. View

Cupples L, Bailey J, Cartier K, Falk C, Liu K, Ye Y . Data mining. Genet Epidemiol. 2005; 29 Suppl 1:S103-9. DOI: 10.1002/gepi.20117. View

Huang T, Kecman V . Gene extraction for cancer diagnosis by support vector machines--an improvement. Artif Intell Med. 2005; 35(1-2):185-94. DOI: 10.1016/j.artmed.2005.01.006. View

Cole S, Hernan M . Fallibility in estimating direct effects. Int J Epidemiol. 2002; 31(1):163-5. DOI: 10.1093/ije/31.1.163. View

Motsinger-Reif A, Reif D, Fanelli T, Ritchie M . A comparison of analytical methods for genetic association studies. Genet Epidemiol. 2008; 32(8):767-78. DOI: 10.1002/gepi.20345. View

Robins J, Greenland S . Identifiability and exchangeability for direct and indirect effects. Epidemiology. 1992; 3(2):143-55. DOI: 10.1097/00001648-199203000-00013. View

10.

van der Laan M . Targeted maximum likelihood based causal inference: Part I. Int J Biostat. 2011; 6(2):Article 2. PMC: 3126670. View

11.

Nonyane B, Foulkes A . Multiple imputation and random forests (MIRF) for unobservable, high-dimensional data. Int J Biostat. 2012; 3(1):Article 12. DOI: 10.2202/1557-4679.1049. View

12.

Bureau A, Dupuis J, Falls K, Lunetta K, Hayward B, Keith T . Identifying SNPs predictive of phenotype using random forests. Genet Epidemiol. 2004; 28(2):171-82. DOI: 10.1002/gepi.20041. View

13.

Ge D, Zhu H, Huang Y, Treiber F, Harshfield G, Snieder H . Multilocus analyses of Renin-Angiotensin-aldosterone system gene variants on blood pressure at rest and during behavioral stress in young normotensive subjects. Hypertension. 2006; 49(1):107-12. DOI: 10.1161/01.HYP.0000251524.00326.e7. View

14.

Lunn D, Whittaker J, Best N . A Bayesian toolkit for genetic association studies. Genet Epidemiol. 2006; 30(3):231-47. DOI: 10.1002/gepi.20140. View

15.

Tannenbaum S, Holford N, Lee H, Peck C, Mould D . Simulation of correlated continuous and categorical variables using a single multivariate distribution. J Pharmacokinet Pharmacodyn. 2006; 33(6):773-94. DOI: 10.1007/s10928-006-9033-1. View

16.

Segal M, Barbour J, Grant R . Relating HIV-1 sequence variation to replication capacity via trees and forests. Stat Appl Genet Mol Biol. 2006; 3:Article2. DOI: 10.2202/1544-6115.1031. View

17.

Hernan M, Hernandez-Diaz S, Werler M, Mitchell A . Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. Am J Epidemiol. 2002; 155(2):176-84. DOI: 10.1093/aje/155.2.176. View

18.

Shohet R, Vega G, Bersot T, Mahley R, Grundy S, Guerra R . Sources of variability in genetic association studies: insights from the analysis of hepatic lipase (LIPC). Hum Mutat. 2002; 19(5):536-42. DOI: 10.1002/humu.10079. View

19.

Diaz-Uriarte R, de Andres S . Gene selection and classification of microarray data using random forest. BMC Bioinformatics. 2006; 7:3. PMC: 1363357. DOI: 10.1186/1471-2105-7-3. View

20.

Costello T, Falk C, Ye K . Data mining and computationally intensive methods: summary of Group 7 contributions to Genetic Analysis Workshop 13. Genet Epidemiol. 2003; 25 Suppl 1:S57-63. DOI: 10.1002/gepi.10285. View