Make a Sparse GRM file

PLINK binary files with high-quality genotyped variants are required to make a sparse GRM. For example, we used 227,437 LD pruned, high-quality genetic variants (MAF > 0.05 and missing rate < 0.05) to calculate sparse GRM and IBD probibalities in SPAGRM paper. Use the command below to extract high-quality genetic variants in PLINK:

--maf 0.05
--geno 0.05
--indep-pairwise 500 50 0.2

Before next step, you should have below PLINK binary files:

XXX.bed,   XXX.bim,   XXX.fam (XXX should be the same prefix of the PLINK binary files)

Run getTempFilesFullGRM to get temporary files 

getTempFilesFullGRM(PlinkFile,
                    nPartsGRM,
                    partParallel,
                    tempDir = NULL,
                    subjData = NULL,
                    minMafGRM = 0.01,
                    maxMissingGRM = 0.1,
                    threadNum = 8)

  • PlinkFile: required. A path to PLINK files (without file extensions of bed/bim/fam).

  • nPartsGRM: required. A numeric value (e.g. 250) to split subjects to multiple parts. For UK Biobank data analysis with ~500K samples, it is recommended to set nPartsGRM=250.

  • partParallel: required. A numeric value (from 1 to partParallel) to split all jobs for parallel computation.

  • tempDir: optional. A path to store temp files to be passed to getSparseGRM. Default is system.file("SparseGRM", "temp", package = "GRAB"). If the sample size > 100K, then the temporary files might need a large amount of space. This should be consistent to the input of getSparseGRM!

  • subjData: optional. A character vector to specify subject IDs to retain (i.e. IID). Default is NULL, i.e. all subjects are retained in sparse GRM.

  • minMafGRM: optional. Minimal value of MAF cutoff to select markers (from PLINK files) to make sparse GRM. (default=0.01)

  • maxMissingGRM: optional. Maximal value of missing rate to select markers (from PLINK files) to make sparse GRM. (default=0.1)

  • threadNum: optional. Number of threads (CPUs) to use.

Example:

GenoFile = system.file("extdata", "simuPLINK.bed", package = "GRAB")
PlinkPrefix = tools::file_path_sans_ext(GenoFile)   # remove file extension
nPartsGRM = 2
for(partParallel in 1:nPartsGRM)
{
   getTempFilesFullGRM(PlinkPrefix, 
                       nPartsGRM = nPartsGRM, 
                       partParallel = partParallel)
}

Run getSparseGRM to combine the temporary files 

getSparseGRM(PlinkFile,
             nPartsGRM,
             SparseGRMFile,
             tempDir = NULL,
             relatednessCutoff = 0.05,
             minMafGRM = 0.01,
             maxMissingGRM = 0.1,
             rm.tempFiles = FALSE)

  • PlinkFile: required. A path to PLINK files (without file extensions of bed/bim/fam). It should be the same as used in getTempFilesFullGRM function.

  • nPartsGRM: required. A numeric value (e.g. 250) to split subjects to multiple parts. For UK Biobank data analysis with ~500K samples, it is recommended to set nPartsGRM=250. It should be the same as used in getTempFilesFullGRM function.

  • SparseGRMFile: required. A file path to store sparse GRM. (e.g. XXX.txt)

  • tempDir: optional. A path to store temp files from getTempFilesFullGRM. Default is system.file("SparseGRM", "temp", package = "GRAB"). This should be consistent to the input of getTempFilesFullGRM!

  • relatednessCutoff: optional. A cutoff for sparse GRM, only GRM elements greater than this cutoff will be retained in sparse GRM. (default=0.05)

  • minMafGRM: optional. Minimal value of MAF cutoff to select markers (from PLINK files) to make sparse GRM. (default=0.01)

  • maxMissingGRM: optional. Maximal value of missing rate to select markers (from PLINK files) to make sparse GRM. (default=0.1)

  • rm.tempFiles: optional. A logical value indicating if the temp files generated in getTempFilesFullGRM will be deleted. (default=FALSE)

Example:

GenoFile = system.file("extdata", "simuPLINK.bed", package = "GRAB")
PlinkPrefix = tools::file_path_sans_ext(GenoFile)   # remove file extension
nPartsGRM = 2
SparseGRMFile = system.file("SparseGRM", "SparseGRM.txt", package = "GRAB")
getSparseGRM(PlinkPrefix, 
             nPartsGRM = nPartsGRM, 
             SparseGRMFile = SparseGRMFile,
             relatednessCutoff = 0.05)

About the SparseGRMFile

The below gives more details about the SparseGRMFile:

SparseGRMFile = system.file("SparseGRM", "SparseGRM.txt", package = "GRAB")
SparseGRM = data.table::fread(SparseGRMFile)
print(SparseGRM)
#            ID1      ID2     Value
#    1:     f1_1     f1_1 0.9550625
#    2:     f1_2     f1_2 1.0272297
#    3:     f1_3     f1_3 1.0192574
#    4:     f1_4     f1_4 1.0053836
#    5:     f1_5     f1_1 0.4648096
#   ---
# 2547: Subj-496 Subj-496 1.0027448
# 2548: Subj-497 Subj-497 0.9913247
# 2549: Subj-498 Subj-498 0.9785985
# 2550: Subj-499 Subj-499 1.0109795
# 2551: Subj-500 Subj-500 0.9783296

SparseGRMFile will latter be used for calculating IBD probabilities.

Note

  • GRAB package implicitly uses GCTA software to make the sparse GRM. As required by GCTA software, the function getSparseGRM is only supported in Linux operation system and PLINK binary files with the same prefix are required.
  • If users use other software (e.g. GCTA) to calculate a sparse GRM file, please convert it to the form above. The column names of sparseGRM file should be exactly ID1, ID2 and Value, separated by a tabulation (\t)!
  • User can also refer to GRAB to make a sparse GRM file.