################################################################################
################################# TA10 - HLA Typing ############################
################################################################################

load("hladb.RData")

samples = hladb$Sample
hamming = data.frame()
sample1 = hladb[which(hladb$Sample == "Recipient"),-1]

for(i in 2:length(samples)) {
  sample2 = hladb[which(hladb$Sample == samples[i]),-1]
  
  hammingDistance = sum(sample1 != sample2) # This is how we ham

  hamming = rbind(hamming, data.frame(samples[i], hammingDistance))
}
colnames(hamming) = c("Sample2", "Hamming")

# We need at 10 matches so we say it's a match
# But Hamming measures DISTANCE and not SIMILARITY
# So let's calculate the maximum hamming distance for it to be a match:
# We have 19 measurements, out of which we need 10 to match - meaning we can have
# a maximum distance of 19-10 = 9

hamming$Sample2[hamming$Hamming <= 9]