Multivariate set-Based Association Test (mBAT), a novel gene-based method, to improve power of identifying genes harbouring variants with masking effects. We show by extensive simulations that mBAT has greater power than the sum-χ^2 strategy (taking fastBAT and MAGMA as examples) in the presence of masking effects. To maximise overall power regardless of the masking effects, we further proposed a hybrid method, mBAT-combo, by combining mBAT and fastBAT test statistics through a Cauchy combination method, a recently developed method to combine different test statistics without knowing the correlation structure(Liu Y, Chen S, Li Z, Morrison AC, Boerwinkle E, Lin X. ACAT: A Fast and Powerful p Value Combination Method for Rare-Variant Analysis in Sequencing Studies. Am J Hum Genet. 2019 Mar 7;104(3):410-421. doi: 10.1016/j.ajhg.2019.01.002. PMID: 30849328; PMCID: PMC6407498.).
You can install the development version of mBAT from GitHub with:
# install.packages("devtools")
devtools::install_github("Share-AL-work/mBAT.combo")This is a basic example which shows you how to solve a common problem:
library(mBAT)path <- system.file("extdata",package = "mBAT")
list.files(path)
#> [1] "1000G_eur_unrel_ukbcom22.bim" "Chr22.fastGWA.ma"
#> [3] "fb_Chr22_1kg.gene.fastbat" "hg19_v40_glist_Pat_v1.txt"
#> [5] "LD_Chr22" "mBAT_R_Chr22.txt"
bim_file <- paste0(path,"/","1000G_eur_unrel_ukbcom22.bim")
map_file <- paste0(path,"/","hg19_v40_glist_Pat_v1.txt")
assoc_file <- paste0(path,"/","Chr22.fastGWA.ma")
LD_path <- paste0(path,"/","LD_Chr22")
#result_path <- paste0(path,"/","extdata")
fastBAT_output <- paste0(path,"/","fb_Chr22_1kg.gene.fastbat")prop=0.9
#res <- mBAT_combo(bim_file,
# map_file,
# assoc_file,
## LD_path,
# result_path,
# fastBAT_output,
# prop,
# gene_annotate)
#head(res)