calcMAF: Takes a recoded (0,1,2) data.table and returns a two column...

In gziegler/ionomicsUtils: Utility functions for ionomics data analysis

Description

Takes a recoded (0,1,2) data.table and returns a two column data.table of num missing and minor allele frequency ((minor alelle)/total) FracMissing is number of NA values/number of lines MAF is (number of minor allele + 0.5 the number of het allele)/(number of nonNA alleles)

Usage

calcMAF(genoTable)

Arguments

genoTable

Data table where SNPs are rows and lines are columns, no metadata columns. Coded as 0,1,2.

Value

Returns a data.table with two columns and rows in the same order as the input. One column contains the MAF and one contains the FracMissing for each row.

Author(s)

Greg Ziegler

Examples

alleleTable <- cbind(alleleTable,calcMAF(genoTable))
#decide who to delete
alleleTable[,discard := (FracMissing>0.2 | MAF<0.1)]

#get rid of the rows in snp that are TRUE in alleleTable$discard
#memory efficient approach to removing row
# see http://stackoverflow.com/questions/10790204/how-to-delete-a-row-by-reference-in-data-table?lq=1
keepIdxs <- which(alleleTable$discard==FALSE)
cols <- names(GenotypeData)
firstCol <- cols[1]
snp.subset <- data.table(col1 = GenotypeData[[firstCol]][keepIdxs]) 
setnames(snp.subset,"col1",firstCol)
for(col in cols[2:length(cols)]){
  snp.subset[, (col) := GenotypeData[[col]][keepIdxs],]
  GenotypeData[, (col):= NULL,] #delete
}
rm(GenotypeData)
snpInfo <- alleleTable[discard==FALSE]

gziegler/ionomicsUtils documentation built on June 20, 2019, 8:04 p.m.