buckley@reed.UUCP (Ken Buckley) (07/22/90)
I have a question for the numerically-minded geneticists out there. What I would like to know is how large the human population would have to be for there to be a statistically significant chance that visually indistinguishable "twins" exist. Naturally I don't care about minor differences in internal organs or tiny details like number of hairs; just another unrelated person who at a pretty close examination looks just like me. Of course the answer is huge but I can't imagine it's *too* much more than 5x10^9 -- e.g. the man in NY who looks just like the Zodiac killer. Is around 10^18 high enough? Please respond by email -- I'll summarize if interest seems to warrant. Thanx. Ken Buckley, buckley@reed.UUCP
pb1t+@andrew.cmu.edu (Peter B. Berget) (07/23/90)
One could begin to make such an estimate by asking what is the probability that non-twin siblings (from the same parents) could be genetically identical. This number is quite staggering in its own right! Humans have 46 chromosomes (22 pairs of autosomes and one pair of sex chromosomes) in their cells. These cells are "diploid" and are designated as having 2N chromosomes, where N is the "haploid" number of chromosomes. During the production of gametes (sperm and eggs) which occurs in the gonads through a cellular process called meiosis, the chromosomes are randomly (!!) divided up so that any gamete will have one of each of the 22 autosomes and one sex chromosome. Thus gametes are N, or haploid, in their chromosome content. The number of different combinations possible when meiosis packages chromosomes into gametes by this process of independent assortment is 2^N. The number of possible combinations of maternal or paternal chromosomes in the resulting gametes is 2^23 or ca. 8 million! Thus each gamete that a human produces contains one of 8 million possible assortments of chromosomes inherited from that individual's mother and father. Because fertilization (the union of an egg and sperm to regenerate a diploid cell which developes into an individual) is random, the chance that two siblings carry the exact same set of chromosomes from their parents is 1/[(2^23)*(2^23)] or 1/(7.04x10^13) to be more accurate. It is no wonder siblings can be so different! Also you should consider that genetically inherited visible traits are probably scattered throughout the 23 different chromosomes. Furthermore each different set of parents generates another 70 trillion, or so, chromosome combinations from the set of chromosomes which they inherited from their parents ........ etc. This estimate of genetic "uniqueness" of each individual takes only into account the random "chromosomal mechanics" which occurs during meiosis and fertilization. It does not take into account "genetic recombination" in which pairs of chromosomes can "swap" genetic information during the first step of meiosis. This would further complicate the process making an estimate of the probability of an individual generating identical gametes. I know you did not ask about genetic identity but this should be a starting point. Peter Berget Department of Biological Sciences Carnegie Mellon University
werner@aecom.yu.edu (Craig Werner) (07/23/90)
In article <15227@reed.UUCP>, buckley@reed.UUCP (Ken Buckley) writes: > I have a question for the numerically-minded geneticists out there. What I > would like to know is how large the human population would have to be for > there to be a statistically significant chance that visually indistinguishable > "twins" exist. Big. Really big. Astronomically large. Now as for visually virtually indistinguishable, or at least close enough to draw stares, I point out the second 'Separated at Birth?' volume is currently available in paperback. -- Craig Werner (future MD/PhD, 5.5 years down, 2.5 to go) werner@aecom.YU.EDU -- Albert Einstein College of Medicine (1935-14E Eastchester Rd., Bronx NY 10461, 212-931-2517) "Disinformation is one thing, but misinformation is unforgiveable."