MHC HAPLOTYPE ANALYSIS BY ARTIFICIAL NEURAL NETWORKS

Conventional matching is based on numbers of alleles shared between do nor and recipient. This approach, however, ignores the degree of relat ionship between alleles and haplotypes, and therefore the actual degre e of difference. To address this problem, we have compared family memb ers using a block matching technique which reflects differences in gen omic sequences. All parents and siblings had been genotyped using conv entional MHC typing so that haplotypes could be assigned and relatives could be classified as sharing 0, 1 or 2 haplotypes. We trained an Ar tificial Neural Network (ANN) with subjects from 6 families (85 compar isons) to distinguish between relatives. Using the outputs of the ANN, we developed a score, the Histocompatibility Index (HI), as a measure of the degree of difference. Subjects from a further 3 families (106 profile comparisons) were rested. The HI score for each comparison was plotted. We show that the HI score is trimodal allowing the definitio n of three populations corresponding to approximately 0, 1 or 2 haplot ype sharing. The means and standard deviations of the three population s were found. As expected, comparisons between family members sharing 2 haplotypes resulted in high HI scores with one exception. More incre asingly, this approach distinguishes between the 1 and 0 haplotype gro ups, with some informative exceptions. This distinct ion was considere d too difficult to attempt visually. The approach provides promise in the quantification of degrees of histocompatibility. (C) American Soci ety for Histocompatibility and Immunogenetics, 1998. Published by Else vier Science Inc.