mergeReplicates {PolyHaplotyper}

merge replicate samples in dosage matrix

Description

merge replicate samples in dosage matrix

Usage

mergeReplicates(mrkDosage, replist, solveConflicts=TRUE)

Arguments

mrkDosage

a dosage matrix with markers in rows and individuals in columns. row names are marker names, column names are individual names.

replist

a list of character vectors, each of which has the sample names of a set of replicates

solveConflicts

if TRUE (default) and there are conflicting dosage assignments between replicates for the same marker, the one with highest frequency is used, provided the total freq of other dosages = 1 OR <= 10% of the frequency of the most frequent dosage. If solveConflicts is FALSE and there are conflicting dosages, the consensus for that marker will be NA

Details

This function merges all sets of replicates, each to one column. The column name of the one retained column is the first one for that set in replist.
For each set of replicates it calls getConsensusmrkDosage

Value

a version of mrkDosage in which only one column of each set of replicates is retained; this column (the first in its set as specified in replist) now has the consensus scores over all replicates. Also, if mrkDosage was a data.frame, it is converted into a matrix.

Examples

# construct a dosage matrix with some missing data:
dosmat <-
 matrix(c(rep(c(3,0,1),3), rep(c(1,1,2),4)), nrow=3,
        dimnames=list(c("mrk1","mrk2","mrk3"),
                      c("a1","a2","a3", "b1","b2","b3","b4")))
ix <- matrix(c(1,1, 3,1, 2,2, 1,3, 2,4, 1,5, 2,6), ncol=2, byrow=TRUE)
dosmat[ix] <- NA
dosmat
# define 2 sets of replicates:
reps <- list(c("a1","a2","a3"), c("b1","b2","b3","b4"))
# merge:
mergeReplicates(mrkDosage=dosmat, replist=reps)
# introduce a conflicting dosage:
dosmat[3,2] <- 2
# merge:
mergeReplicates(mrkDosage=dosmat, replist=reps)