USING NEURAL NETWORKS AS AN AID IN THE DETERMINATION OF DISEASE STATUS - COMPARISON OF CLINICAL-DIAGNOSIS TO NEURAL-NETWORK PREDICTIONS IN A PEDIGREE WITH AUTOSOMAL-DOMINANT LIMB-GIRDLE MUSCULAR-DYSTROPHY

Studies of the genetics of certain inherited diseases require expertis e in the determination of disease status even for single-locus traits. For example, in the diagnosis of autosomal dominant limb-girdle muscu lar dystrophy (LGMD1A), it is not always possible to make a clear-cut determination of disease, because of variability in the diagnostic cri teria, age at onset, and differential presentation of disease. Mapping such diseases is greatly simplified if the data present a homogeneous genetic trait and if disease status can be reliably determined. Here, we present an approach to determination of disease status, using meth ods of artificial neural-network analysis. The method entails ''traini ng'' an artificial neural network, with input facts (based on diagnost ic criteria) and related results (based on disease diagnosis). The net work contains weight factors connecting input ''neurons'' to output '' neurons,'' and these connections are adjusted until the network can re liably produce the appropriate outputs for the given input facts. The trained network can be ''tested'' with a second set of facts, in which the outcomes are known but not provided to the network, to see how we ll the training has worked. The method was applied to members of a ped igree with LGMD1A, now mapped to chromosome 5q. We used diagnostic cri teria and disease status to train a neural network to classify individ uals as ''affected'' or ''not affected.'' The trained network reproduc ed the disease diagnosis of all individuals of known phenotype, with 9 8% reliability. This approach defined an appropriate choice of clinica l factors for determination of disease status. Additionally, it provid ed insight into disease classification of those considered to have an ''unknown'' phenotype on the basis of standard clinical diagnostic met hods.