ki.db: Computes KIs for case profile(s) with all database profiles
In DNAprofiles: DNA Profiling Evidence Analysis

Computes KIs for case profile(s) with all database profiles

1
2
3

ki.db(x, db, hyp.1, hyp.2 = "UN", freqs = get.freqs(x),
  markers = intersect(get.markers(x), get.markers(db)), theta = 0,
  ret.per.marker = FALSE, precomputed.kis)

`x`	An integer matrix specifying either a single profile or a number of profiles. Alternatively an integer vector containing a single profile, e.g. obtained when a row is selected from a matrix of profiles.
`db`	An integer matrix which is the database of profiles.
`hyp.1`	A character string giving the hypothesis in the numerator of the KI. Should be one of ibdprobs, e.g. "FS" (full sibling) or "PO" (parent/offspring) or "UN" (unrelated).
`hyp.2`	A character string giving the hypothesis in the denominator of the KI. Should be one of ibdprobs, e.g. "FS" (full sibling) or "PO" (parent/offspring) or "UN" (unrelated). Defaults to "UN".
`freqs`	A list specifying the allelic frequencies. Should contain a vector of allelic frequencies for each locus, named after that locus.
`markers`	Character vector stating the markers to use in the KI computation. Defaults to the intersection of the markers of `x1` and `x2`.
`theta`	Numeric value specifying the amount of background relatedness.
`ret.per.marker`	Logical. If TRUE, return a matrix of KIs, where the columns correspond to markers.
`precomputed.kis`	(optionally) a list of precomputed KIs, returned by `ki.precompute`. This speeds up the computation when multiple profiles are run against the db (i.e. `x` has more than one row).

data(freqsNLsgmplus)
fr <- freqsNLsgmplus

# sample a profile, a database and compute the Sibling Index (SI) with all database members
x <- sample.profiles(N=1,fr)
db <- sample.profiles(N=10^4,fr)
si <- ki.db(x,db=db,"FS")

# estimate the exceedance probabilities of an SI-threshold
t <- 1 # choose threshold SI=1
x <- sample.profiles(N=1,fr)
sibs <- sample.relatives(x,N=10^4,type="FS")
unrelated <- sample.profiles(N=10^4,fr)
mean(ki.db(x,db=sibs,"FS")>=t) # the vast majority of true siblings has an SI>=1
mean(ki.db(x,db=unrelated,"FS")>=t) # a few percent of unrelated persons have SI >= 1

# estimate distribution of SI for true siblings and unrelated persons
x <- sample.profiles(N=1,fr) #sample profile
sibs <- sample.relatives(x,N=10^4,type="FS") #sample sibs
unrelated <- sample.profiles(N=10^4,fr) #sample unrelated persons

sibs.si <- ki.db(x,db=sibs,"FS") #compute SI for true siblings
unrelated.si <- ki.db(x,db=unrelated,"FS") #compute SI for unrelated persons
#plot density estimates of SI
plot(density(log10(unrelated.si)),xlim=c(-10,10),lty="dashed",
    xlab=expression(log[10](SI)),main="SI for true sibs and unrelated profiles")
lines(density(log10(sibs.si)))