As the focus of genome-wide scans for disease loci have shifted from simple
Mendelian traits to genetically complex traits, researchers have begun to
consider new alternative ways to detect linkage that will consider more tha
n the marginal effects of a single disease locus at a time, One interesting
new method is to train a neural network on a genome-wide data set in order
to search for the best non-linear relationship between identity-by-descent
sharing among affected siblings at markers and their disease status. We in
vestigate here the repeatability of the neural network results from run to
run, and show that the results obtained by multiple runs of the neural netw
ork method may differ quite a bit, This is most likely due to the fact that
training a neural network involves minimizing an error function with a mul
titude of local minima. Copyright (C) 2001 S. Karger AG, Basel.