RESUMEN
Empirical analysis of 200 paternity cases with the F10 multilocus DNA fingerprinting probe showed that it was capable of distinguishing fathers from nonfathers in every case. The average exclusion probability was0.99998. A very effective discrimination parameter was the proportion of nonmaternal (test) bands which cannot be detected in the alleged father (unassignable bands) among all test bands. Values below 0.2 were seen in true fathers while in all cases of nonfathers the values were above 0.35. Minisatellite mutations occurred at a rate of 0.004 per band per child. The distribution of band-sharing among first degree relatives and unrelated individuals showed only a small overlap. Thus, band-sharing of the F10 fingerprints should provide a useful statistics for testing genetic relationships in deficiency cases.