HUMAN-CHROMOSOME VARIATION - THE DISCRIMINATORY POWER OF Q-BAND HETEROMORPHISM (VARIANT) ANALYSIS IN DISTINGUISHING BETWEEN INDIVIDUALS, WITH SPECIFIC APPLICATION TO CASES OF QUESTIONABLE PATERNITY

  • 1 February 1986
    • journal article
    • research article
    • Vol. 38  (2) , 235-252
Abstract
The chromosomes from 57 persons were analyzed by means of quinacrine fluorescent staining in order to assess the amount of variation and the discriminatory power of Q-band heteromorphism analysis. Chromosomes 3, 4, 13, 14, 15, 21, 22, and Y of each person were visually compared to those of 56 others, for a total of 1,596 comparisons. No two persons were found to have the same set of variants. The number of differences between chromosomes for each comparison ranged from 2 to 12 out of a possible total of 14 for females and 15 for males. Relatives were also distinguishable, and differences ranged from two to seven. We used the frequency with which each chromosome was useful for telling two people apart, and estimated the probability of finding two persons with the same set of quinacrine variants as .0003. Distinctly different heteromorphisms were found in the 39 unrelated persons for each of the chromosomes examined. In this small population, the number of different sets of variants observed for chromosomes 3, 4, 13, 14, 15, 21, 22, and Y were six, seven, 27, 16, 20, 15, 24, and five, respectively, for a total number of possible combinations of 1.14 .times. 1015. As a test of the usefulness of chromosome heteromorphisms in paternity cases, 12 father-mother-child trios of virtually certain paternity, owing to the father-child segregation of a rare structural rearrangement, were coded and recombined at random to produce 120 cases of uncertain paternity. When the code was broken, 108 "alleged fathers" had been excluded correctly and the 12 biological fathers had been included correctly. Variants for the 39 unrelated persons were scored according to a modified version of the system proposed at the Paris Conference. We calculated the probability of excluding the wrongfully accused man for each chromosome based on the frequencies of specific variants found in this population. The calculations took into consideration not only the observed sets of variants, but also all possible combinations of the specific variants. The individual probability of exclusion for each chromosome was then used to calculate a cumulative probability of exclusion for all of these chromosomes of 1.0000.