Improving power and accuracy of genome-wide association studies via a multi-locus mixed linear model methodology

Abstract

Genome-wide association studies (GWAS) have been widely used in genetic dissection of complex traits. However, common methods are all based on a fixed-SNP-effect mixed linear model (MLM) and single marker analysis, such as efficient mixed model analysis (EMMA). These methods require Bonferroni correction for multiple tests, which often is too conservative when the number of markers is extremely large. To address this concern, we proposed a random-SNP-effect MLM (RMLM) and a multi-locus RMLM (MRMLM) for GWAS. The RMLM simply treats the SNP-effect as random, but it allows a modified Bonferroni correction to be used to calculate the threshold p value for significance tests. The MRMLM is a multi-locus model including markers selected from the RMLM method with a less stringent selection criterion. Due to the multi-locus nature, no multiple test correction is needed. Simulation studies show that the MRMLM is more powerful in QTN detection and more accurate in QTN effect estimation than the RMLM, which in turn is more powerful and accurate than the EMMA. To demonstrate the new methods, we analyzed six flowering time related traits in Arabidopsis thaliana and detected more genes than previous reported using the EMMA. Therefore, the MRMLM provides an alternative for multi-locus GWAS.

Analysis of QTLs on heading date based on single segment substitution lines in rice (Oryza Sativa L.)

“…S1) 11 . Analysis of molecular markers (more than 460 SSR) confirmed that each SSSL contained single substituted segment from a donor only, and all substitution segments of SSSLs were distributed on 12 chromosomes with the average size of 15.8 cM and the coverage rate of 97% over the whole rice genome 13 .…”

Genome wide association study in 3,173 outbred rats identifies multiple loci for body weight, adiposity, and fasting glucose

“…GWAS analysis employed a linear mixed model, as implemented in the software GEMMA (Wang et al 2016), using dosages and a genetic relatedness matrix (GRM) to account for the complex family relationships within the HS population. We used the Leave One Chromosome Out (LOCO) method to guard against proximal contamination, as previously described Gonzales et al 2018).…”

mrMLM v4.0: An R Platform for Multi-locus Genome-wide Association Studies

“…Using mrMLM v4.0, individual parameters may be changed in order to obtain the best results (Files S5 and S6). For example, the number of potentially associated SNPs for each chromosome in Zhang et al [20] is set at 50, and the search radius in mrMLM [16] and FASTmrMLM [17] is set at 20 kb in real data analysis. In addition, users should understand some parameter settings.…”

“…Thereafter, Liu et al [14] developed FarmCPU. Based on the advantages of the random model of QTN effect over the fixed model [15], recently, we have developed six multi-locus methods: mrMLM [16], FASTmrMLM [17], FASTmrEMMA [18], ISIS EM-BLASSO [19], pLARmEB [20] and pKWmEB [21] (Files S1 and S2). These methods include two stages.…”