Package: paneljudge 0.0.0.9000

paneljudge: Judge the performance of a panel of genetic markers using simulated data

An R package to judge the performance of a panel of genetic markers using data simulated for pairs of haploid genotypes. The data are simulated under a hidden Markov model of relatedness (described in Taylor, A.R., Jacob, P.E., Neafsey, D.E. and Buckee, C.O., 2019. Estimating relatedness between malaria parasites. Genetics, 212(4), pp.1337-1351) using allele frequency estimates provided by the user and inter-marker distances. The markers are treated as categorical random variables whose realisations (alleles) are unordered. The effective cardinalities and diversities of the markers can be computed using the input allele frequency estimates. Panel performance can be judged in terms of the root mean square error (RMSE) and confidence interval width of estimated relatedness, where relatedness is estimated under the same model used to simulate the data. At present, the examples we provide do not consider model misspecification; do not account for uncertainty around input allele frequency estimates; do not consider relatedness between pairs of haploid genotypes simulated using different allele frequencies; do not account for marker drop-out (markers that fail to produce useful data, e.g. because they a monomorphic). Otherwise stated, in the examples provided, the performance of a panel is judged in its most favourable light; it will likely perform less well in reality.

Authors:Aimee Taylor [aut, cre], Pierre Jacob [aut]