An efficient multi-locus mixed model approach for genome-wide association studies in structured populations

By V. Segura, Bjarni J. Vilhjálmsson, Alexander Platt, Arthur Korte, Ümit Seren, Quan Long, Magnus Nordborg. Nature Genetics, 2012

955 citations recorded by the source corpus.

Population structure causes genome-wide linkage disequilibrium between unlinked loci, leading to statistical confounding in genome-wide association studies. Mixed models have been shown to handle the confounding effects of a diffuse background of large numbers of loci of small effect well, but they do not always account for loci of larger effect. Here we propose a multi-locus mixed model as a general method for mapping complex traits in structured populations. Simulations suggest that our method outperforms existing methods in terms of power as well as false discovery rate. We apply our method to human and Arabidopsis thaliana data, identifying new associations and evidence for allelic heterogeneity. We also show how a priori knowledge from an A. thaliana linkage mapping study can be integrated into our method using a Bayesian approach. Our implementation is computationally efficient, making the analysis of large data sets (n > 10,000) practicable.

Subfields: Genetics

View Paper: https://www.semanticscholar.org/paper/b0a8003939b352cae26ce97d235881c213915093

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Genetics

Citations in the corpus

Iterative Usage of Fixed and Random Effect Models for Powerful and Efficient Genome-Wide Association Studies (2016, methodology): “These methods are: (1) naïve method (t-test); (2) GLM[8]; (3) MLM[9,26]; (4) CMLM[17]; (5) FaST-LMM-Select[23], and (6) MLMM[25].”
Polygenic Modeling with Bayesian Sparse Linear Mixed Models (2012, methodology): “Indeed, models like these have been used in breeding value prediction to assist genomic selection in animal and plant breeding programs [30–35], gene selection in expression analysis while controlling for batch effects [36], phenotype prediction of complex traits in model organisms and dairy cattle ”
Variable Selection in Heterogeneous Datasets: A Truncated-rank Sparse Linear Mixed Model with Applications to Genome-wide Association Studies (2017, background): “Interest in these methods has surged recently given improvements that allow their application to human-scale genome data [16], [17], [18], [19], [20], [21].”
Improving power and accuracy of genome-wide association studies via a multi-locus mixed linear model methodology (2016, neutral): “Finally, the MRMLM method uses one set of selected SNPs, which have p values less than 0.01 in the initial scanning while Segura et al. 13 requires MCMC samplings.”
The advantages and limitations of trait analysis with GWAS: a review (2013, background): “It becomes possible to disentangle the contribution of genetic heterogeneity by including ‘competing’ variants as cofactors within a mixed model setting [39].”
Genetic mapping of quantitative trait loci in crops (2017, background): “[63] shows promising performance in terms of false positives and power, using a”
rMVP: A Memory-efficient, Visualization-enhanced, and Parallel-accelerated Tool for Genome-wide Association Study (2020, methodology): “Multiple algorithms have been developed to 50 boost both the computational efficiency and statistical power of MLM methods[8-11].”
Genome‐wide Association for Plant Height and Flowering Time across 15 Tropical Maize Populations under Managed Drought Stress and Well‐Watered Conditions in Sub‐Saharan Africa (2016, background): “Model selection improves GWAS results by identifying groups of individual SNPs that best explain the variance in the model (e.g., Wang et al. 2011; Segura et al. 2012).”
GAPIT Version 2: An Enhanced Integrated Tool for Genomic Association and Prediction (2016, methodology): “The software package also did not implement the functions to derive kinship from both pedigree and genetic markers (Misztal et al., 2009), models of multiple traits (Arthur et al., 2012), and multiple marker tests (Segura et al., 2012).”
BLINK: a package for the next level of genome-wide association studies with both individuals and markers in the millions (2018, neutral): “The third adjustment, known as the multi-locus mixed-model (MLMM) approach, applies elimination of the kinship [28].”
Efficient Bayesian mixed model analysis increases association power in large cohorts (2014, methodology): “To compute the calibration constant cinf in equation (9), BOLT-LMM rapidly computes the prospective statistic χ(2)LMM-LOCO from equation (7) at 30 random SNPs by applying conjugate gradient iteration to compute VLOCOxtest for each of the 30 selected SNPs xtest.”
Natural variations and genome-wide association studies in crop plants. (2014, methodology): “Some additional multiple-regression approaches and nonparametric statistics have also been developed (56, 100).”
Identifying Causal Variants at Loci with Multiple Signals of Association (2014, background): “This assumption is the basis of many recent approaches to estimate heritability (Yang et al. 2011a,b; Speed et al. 2012; Kostem and Eskin 2013) and to correct for population structure in GWAS (Kang et al. 2008; Lippert et al. 2011; Listgarten et al. 2012; Segura et al. 2012; Zhou and Stephens 2012).”
GAPIT Version 3: Boosting Power and Accuracy for Genomic Association and Prediction (2020, methodology): “Packages in the specific category are usually accompanied by the development 428 of new methods, such as MLMM[14], FarmCPU[15], and BLINK[16].”
Marker-Assisted Plant Breeding: Principles and Practices (2015, methodology): “The MLMM can be easily extended to Bayesian analysis as it is based on a linear model (Segura et al. 2012).”