map_snps_to_genes: Map SNPs to their nearby genes

In NathanSkene/MAGMA_Celltyping: Find Causal Cell-Types Underlying Complex Trait Genetics

Map SNPs to their nearby genes

Description

Make two external calls to MAGMA. First use it to annotate SNPs onto their neighbouring genes. Second, use it to calculate the gene level trait association.

Usage

map_snps_to_genes(
  path_formatted,
  genome_build = NULL,
  upstream_kb = 35,
  downstream_kb = 10,
  N = NULL,
  duplicate = c("drop", "first", "last", "error"),
  synonym_dup = c("skip", "skip-dup", "drop", "drop-dup", "error"),
  genome_ref_path = NULL,
  population = "eur",
  genes_only = FALSE,
  storage_dir = tools::R_user_dir("MAGMA.Celltyping", which = "cache"),
  force_new = FALSE,
  version = NULL,
  verbose = TRUE
)

Arguments

path_formatted	Filepath of the summary statistics file (which is expected to already be in the required format). Can be uncompressed or compressed (".gz" or ".bgz").
genome_build	The build of the reference genome ("GRCh37" or "GRCh38"). If NULL, it will be inferred with get_genome_build.
upstream_kb	How many kilobases (kb) upstream of the gene should SNPs be included?
downstream_kb	How many kilobases (kb) downstream of the gene should SNPs be included?
N	What is the N number for this GWAS? That is cases + controls.
duplicate	The duplicate modifier can be used to specify the desired behaviour for dealing with duplicate SNPs in the file, and can be set to one of four values: 'drop', 'first', 'last', and 'error'. When set to 'drop', the corresponding SNP is removed from the analysis entirely. When set to 'first' or 'last', either the first or the last entry for that SNPs in the file is used. When set to 'error', the program terminates if encountering any duplicate SNPs. The default mode is 'duplicate=drop'. Note that SNPs are only checked for duplication if they are present in the genotype data, and if they have a non-missing pvalue (and sample size, if ncol is set). When synonymous SNP IDs have been loaded, different SNP IDs referring to the same SNP are considered duplicates as well. Unless duplicate is set to 'error', a list of duplicate SNPs will be written to the supplementary log file.
synonym_dup	When loading SNP ID synonyms, MAGMA may detect SNP IDs in the genotype data that are synonyms of each other. The synonym-dup modifier for the –bfile flag can be used to specify the desired behaviour for dealing with such SNPs. This modifier can be set to one of four values: 'drop', 'drop-dup', 'skip', 'skip-dup' and 'error'. When set to 'drop', SNPs that have multiple synonyms in the data are removed from the analysis. Conversely, when set to 'skip' the SNPs are left in the data and the synonym entry in the synonym file is skipped. When set to 'drop-dup', for each synonym entry only the first listed in the synonym file is retained; for subsequent SNP IDs in the same entry that are found in the data are removed, and their IDs are mapped as synonyms to the first SNP. When set to 'skipdup' the genotype data for all synonymous SNPs is retained; SNP IDs not found in the data are mapped to the first SNP in the synonym entry that is. Finally, when set to 'error', the program will simply terminate when encountering synonymous SNPs in the data. The default mode is 'synonym-dup=skip'. Unless synonym-dup is set to error, a list of synonymous SNPs in the data will be written to the supplementary log file.
genome_ref_path	Path to the folder containing the 1000 genomes reference (downloaded with get_genome_ref).
population	Which population subset of the genome reference to include. "eur" : European descent (Default simply because this is currently the most common GWAS subpopulation). "afr" : African descent. "amr" : Ad Mixed American descent. "eas" : East Asian descent. "sas" : South Asian descent.
genes_only	The .genes.raw file is the intermediary file that serves as the input for subsequent gene-level analyses. To perform only a gene analysis, with no subsequent gene-set analysis, the --genes-only flag can be added (TRUE). This suppresses the creation of the .genes.raw file, and significantly reduces the running time and memory required.
storage_dir	Where to store genome ref.
force_new	Set to TRUE to rerun MAGMA even if the output files already exist. (Default: FALSE).
version	MAGMA version to use.
verbose	Print messages.

Value

Path to the genes.out file.

Examples

## Not run: 
path_formatted <- MAGMA.Celltyping::get_example_gwas()
genesOutPath <- MAGMA.Celltyping::map_snps_to_genes(
    path_formatted = path_formatted,
    genome_build = "hg19",
    N = 5000)

## End(Not run) 

NathanSkene/MAGMA_Celltyping documentation built on Aug. 21, 2023, 8:55 a.m.