Premium
Application of probabilistic neural network analysis to a disease with complex inheritance: The GAW11 simulated data
Author(s) -
Naimark David M.J.,
Paterson Andrew D.
Publication year - 1999
Publication title -
genetic epidemiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.301
H-Index - 98
eISSN - 1098-2272
pISSN - 0741-0395
DOI - 10.1002/gepi.13701707109
Subject(s) - probabilistic neural network , inheritance (genetic algorithm) , probabilistic logic , artificial neural network , linkage (software) , computer science , statistical model , allele , artificial intelligence , machine learning , genetics , biology , gene , time delay neural network
A probabilistic neural network analysis was applied to the simulated GAW11 data. Six replicates drawn at random from one of the simulated populations were used to generate training and test vectors for pairs of siblings. The vectors incorporated two environmental indicators as well as identical‐by‐descent allele sharing scores from each of 300 genetic markers. The performance of a ‘naïve’ probabilistic neural network (PNN) was fair. However, by combining a traditional linkage analysis with a PNN which incorporated gene×environment interaction, the performance was considerably enhanced.