Preprocess GWAS Summary Statistics for MRcare

This function preprocesses GWAS summary statistics to prepare them for use with MRcare. It standardizes column names, performs quality control, and saves the processed data to a file for later use.

Usage

preprocess_gwas_data(
  gwas_data,
  output_file,
  data_type = c("exposure", "outcome"),
  A1 = NULL,
  A2 = NULL,
  SNP = NULL,
  CHR = NULL,
  POS = NULL,
  BETA = NULL,
  SE = NULL,
  Pval = NULL,
  MAF = NULL,
  N = NULL,
  Zscore = NULL,
  Ninput = NULL,
  maf_filter = 0.01,
  remove_ambiguous = TRUE,
  verbose = TRUE
)

Arguments

gwas_data

Path to a GWAS summary statistics file or a data frame containing GWAS summary statistics

output_file

Path where the processed GWAS data will be saved

data_type

Either "exposure" or "outcome" to determine processing logic

A1

Column name for effect allele (default: NULL, will try to detect)

A2

Column name for other allele (default: NULL, will try to detect)

SNP

Column name for SNP ID (default: NULL, will try to detect)

CHR

Column name for chromosome (default: NULL, will try to detect)

POS

Column name for position (default: NULL, will try to detect)

BETA

Column name for effect size (default: NULL, will try to detect)

SE

Column name for standard error (default: NULL, will try to detect)

Pval

Column name for p-value (default: NULL, will try to detect)

MAF

Column name for minor allele frequency (default: NULL, will try to detect)

N

Column name for sample size (default: NULL, will try to detect)

Zscore

Column name for Z-score (default: NULL, will try to detect)

Ninput

Sample size if not available in data (default: NULL)

maf_filter

Minimum MAF threshold to include SNPs (default: 0.01)

remove_ambiguous

Whether to remove strand-ambiguous SNPs (default: TRUE)

verbose

Whether to print progress messages (default: TRUE)

Value

Path to the processed GWAS data file

Examples

if (FALSE) { # \dontrun{
# Process exposure data
exposure_processed <- preprocess_gwas_data(
  gwas_data = "path/to/raw_exposure_gwas.txt",
  output_file = "path/to/processed_exposure_gwas.rds",
  data_type = "exposure",
  Ninput = 10000
)

# Process outcome data
outcome_processed <- preprocess_gwas_data(
  gwas_data = "path/to/raw_outcome_gwas.txt",
  output_file = "path/to/processed_outcome_gwas.rds",
  data_type = "outcome",
  Ninput = 15000
)
} # }