# Odds of identical DNA

Staff Emeritus
Gold Member
Is it theoretically possible that two people not directly related could be born with identical DNA? I would imagine that there is always the chance of a series of mutations that make one person identical to another, but I have no sense of the magnitude of the odds of such an event - I assume they would be astronomical. I was also wondering if two different sets of parents could theoretically produce children with identical DNA.

JonMoulton
I'll give a shot at considering the problem in a research-free fashion.

Need:
# of bases in the human genome
average % difference in base sequence between siblings

Calculate the average number of bases different. Consider that the odds of randomly matching bases is 1/(4^n) where n is the number of random bases being compared.

Sound like a reasonable approach?

1 person
Gold Member
2022 Award
I have read (sorry, no citations) that although it is technically not IMpossible for two people to have the same DNA, the odds of any given living person having the same DNA as another living person are ... I forget what exact description was used, but your choice of "astronomical" (meaning, I assume, infinitesimal in this case) seems fitting.

1 person
Enigman
Ygg said:
By sequencing the DNA from various individuals, scientists have found 15 million single nucleotide polymorphisms, 1 million short insertions and deletions, and 20,000 structural variants (The 1000 Genomes Project Consortium). Assuming each of these sites of variation has only two alleles and that none are synthetically lethal, this puts an estimate of the possible combinations from randomly assorting these variants at 2^(1.6x10^6) or 10^480000.

1 person
Gold Member
One in 10^480000 would be an upper bound on the probability of two individuals having the same DNA (the calculation was made in the context of estimating the number of possible human genome sequences). For this to be the actual probability would require 1) that these variants are distributed randomly throughout the population with respect to each other (many populations will likely share the same variants), and 2) that the variant and wild-type alleles are present in a 50-50 mixture throughout the population (many variants are likely rare).

Of course, the likelihood of two individuals having identical genomes is still infinitesimally small.

1 person
Homework Helper
Just to complicate things a bit more: if the DNA in a single cell mutates, that mutation doesn't instantly alter every other cell in the body.

So what it the probability that even ONE person has totally "identical" DNA, let alone two people? (The answer obviously depends on the person's age - make whatever assumptions you want.).

Gold Member
I have read (sorry, no citations) that although it is technically not IMpossible for two people to have the same DNA, the odds of any given living person having the same DNA as another living person are ... I forget what exact description was used, but your choice of "astronomical" (meaning, I assume, infinitesimal in this case) seems fitting.

Note: infinitesimal means approaching zero.

Gold Member
2022 Award
Note: infinitesimal means approaching zero.

That's good, since that is exactly the sense in which I used it.

Gold Member
Sorry for the implication, that was intended for the casual reader.

Homework Helper
Gold Member
2022 Award
The odds of two people having the same DNA = the odds that they are clones.

Gold Member
2022 Award
The odds of two people having the same DNA = the odds that they are clones.

Do you have any evidence for that? It doesn't sound right to me. My understanding is that, as I said earlier, it is not technically impossible for two humans to have the same DNA. No cloning is required so the two are unrelated.

Homework Helper
Gold Member
2022 Award
Do you have any evidence for that? It doesn't sound right to me. My understanding is that, as I said earlier, it is not technically impossible for two humans to have the same DNA. No cloning is required so the two are unrelated.

I only meant that, when probabilities get as low as this, then they reduce to the probability that the outcome is achieved not by chance.

I suppose, more precisely, it's:

The probability that two people have the same DNA = the probability that they were artificially engineered that way + the negligible probability that it happened naturally.

Gold Member
2022 Award
I only meant that, when probabilities get as low as this, then they reduce to the probability that the outcome is achieved not by chance.

I suppose, more precisely, it's:

The probability that two people have the same DNA = the probability that they were artificially engineered that way + the negligible probability that it happened naturally.

Well, I disagree with your premise that two numbers deserve the equals sign just because they are both very small. By your reasoning, 10E-30 = 10E-50 and I argue that it does not.

If you want to just say they are both very small, I agree, but your careless use of the equals sign is not helpful.

Enigman
The odds of two people having the same DNA = the odds that they are clones.

Do you have any evidence for that? It doesn't sound right to me. My understanding is that, as I said earlier, it is not technically impossible for two humans to have the same DNA. No cloning is required so the two are unrelated.
Clones, AFAI(&a short google search)K is a term for two genetically identical organisms with no clause about how they came to be genetically identical: artificially, by chance or through asexual reproduction.
Well, I disagree with your premise that two numbers deserve the equals sign just because they are both very small. By your reasoning, 10E-30 = 10E-50 and I argue that it does not.

If you want to just say they are both very small, I agree, but your careless use of the equals sign is not helpful.

His/her premise is rather that 10E-30 +10E-50 ≈ 10E-30 (I think)
10E-30 being analogous to probability of them being artificial clones and 10E-50 being analogous to them having same genetic matter naturally.

Though for the blue to be correct we would have to discount identical twins which is much more likelier than artificial cloning.

Gold Member
2022 Award
His/her premise is rather that 10E-30 +10E-50 ≈ 10E-30 (I think)

Seems to me you are reading between the lines in a way that is not justified by the statement he/she made.

Mentor
And then there's chimeras, people with more than one DNA signature. In one case, a woman had two sets of DNA which was determined to have come from a twin that was somehow absorbed into her shortly after conception.

http://en.wikipedia.org/wiki/Chimera_(genetics)

mafagafo
I was also wondering if two different sets of parents could theoretically produce children with identical DNA.

Yes, and as the similarities between these sets of parents become higher, higher becomes the chance of children with identical DNA being produced.

Two people with the same DNA is already really unlikely, therefore we can go a bit further. But, if these parents have many differences in their DNA we would need more and more mutations for this to work. I don't think this can, mathematically at least, reach a point where it is impossible to have people with the same DNA, but it gets ridiculously unlikely.