likEvid: Likelihood of DNA evidence conditioned on a given hypothesis In forensim: Statistical tools for the interpretation of forensic DNA mixtures

Description

likEvid allows the calculation of likelihood for a piece of DNA evidence, for any number of replicates, any number of contributors, and when drop-in and drop-out are possible.

Usage

 `1` ```likEvid(Repliste, T, V, x, theta, prDHet, prDHom, prC, freq) ```

Arguments

 `Repliste` vector of alleles present at a given locus for any number of replicates. If there are two replicates, showing alleles 12,13, and 14 respectively, then `Repliste` should be given as c(12,13,0,14), where the 0 is used as a separator. An empty replicate is simply 0. For example, replicates (12,13) and and one empty replicate must be given as: c(12,14,0,0). `T` vector of genotypes for the known contributors under Hp. Genotype 12/17 should be given as a vector c(12,17) and genotypes 12/17,14/16, should be given as a unique vector: c(12,17,14,16). If T is empty, set to 0. `V` vector of genotypes for the known non-contributors (see References section) under Hp. See `T` for format. `x` Number of unknown individuals under H. Set to 0 if there are no unknown contributors. `theta` thete correction, value must be taken in [0,1) `prDHet` probability of dropout for heterozygotes. It is possible to assign different values per contributor. In this case, `prDHet` must be a vector, of length the number of contributors in T + x, and the probabilities must be given in this order. if the probability of dropout for T is d1, and for the unknown is d2, then `prDHet`=(d1,d2). In case T is not a heterozygote, the given vector must still be of length length(T) +x, but the given value for T does not matter, because it won't be used, the value in prDHom is used instead. This is a bit ad hoc and an improvement is currently under development. `prDHom` probability of dropout for homozygotes. See description ofr argument `PrDHom`. `prC` probability of drop-in applied per locus `freq` vector of the corresponding allele frequencies of the analysed locus in the target population

Author(s)

Hinda Haned [email protected]

References

Gill, P.; Kirkham, A. & Curran, J. LoComatioN: A software tool for the analysis of low copy number DNA profiles Forensic Science International, 2007, 166(2-3), 128-138

Curran, J. M.; Gill, P. & Bill, M. R. Interpretation of repeat measurement DNA evidence allowing for multiple contributors and population substructure Forensic Science International, 2005, 148, 47-53

`LRmixTK`
 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15``` ```#load allele frequencies library(forensim) data(ngm) #create vector of allele frequencies d10<-ngm\$tab\$D10 # evaluate the evidence under Hp; contributors are the suspect and one unknown, # dropout probabilities for the suspect and the unknown are the same: 0.2 for heterozygotes, # and 0.1 for homozygotes. likEvid(Repliste=c(12,13,14),T=c(12,13),V=0,x=1,theta=0,prDHet=c(0.2,0.2), prDHom=c(0.04,0.04),prC=0, freq=d10) # evaluate the evidence under Hd; contributors are two unknown people, the dropout # probabilities for the unknowns is kept the same under Hd likEvid(Repliste=c(12,13,14),T=0,V=0,x=2,theta=0,prDHet=c(0.2,0.2), prDHom=c(0.04,0.04),prC=0,freq=d10) ```