List of uncorrected column headers often found in GWAS Summary Statistics column headers. Note the effect allele will always be the A2 allele, this is the approach done for VCF(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7805039/). This is enforced with the column header corrections here and also the check allele flipping test.
data("sumstatsColHeaders")dataframe with 2 columns
The code to prepare the .Rda file file from the marker file is:
# Most the data in the below table comes from the LDSC github wiki
data("sumstatsColHeaders")
# Make additions to sumstatsColHeaders using github version of MungeSumstats-
# shown is an example of adding columns for Standard Error (SE)
#se_cols <- data.frame("Uncorrected"=c("SE","se","STANDARD.ERROR",
# "STANDARD_ERROR","STANDARD-ERROR"),
# "Corrected"=rep("SE",5))
#sumstatsColHeaders <- rbind(sumstatsColHeaders,se_cols)
#Once additions are made, order & save the new mapping dataset
#now sort ordering -important for logic that
# uncorrected=corrected comes first
sumstatsColHeaders$ordering <-
sumstatsColHeaders$Uncorrected==sumstatsColHeaders$Corrected
sumstatsColHeaders <-
sumstatsColHeaders[order(sumstatsColHeaders$Corrected,
sumstatsColHeaders$ordering,decreasing = TRUE),]
rownames(sumstatsColHeaders)<-1:nrow(sumstatsColHeaders)
sumstatsColHeaders$ordering <- NULL
#manually move FRWQUENCY to above MAR - github issue 95
frequency <- sumstatsColHeaders[sumstatsColHeaders$Uncorrected=="FREQUENCY",]
maf <- sumstatsColHeaders[sumstatsColHeaders$Uncorrected=="MAF",]
if(as.integer(rownames(frequency))>as.integer(rownames(maf))){
sumstatsColHeaders[as.integer(rownames(frequency)),] <- maf
sumstatsColHeaders[as.integer(rownames(maf)),] <- frequency
}
usethis::use_data(sumstatsColHeaders,overwrite = TRUE, internal=TRUE)
save(sumstatsColHeaders,
file="data/sumstatsColHeaders.rda")
# You will need to restart your r session for effects to take account