It was not 100% -- look it up.
It was 99.9%, which is as much as you can get. It doesn't go to 100. You can have a 100% exclusion (as AA's sample did for the royal family) but as far as proving identity, it doesn't go any higher than 99.9% for anyone.
Some of those 'who's the baby's daddy' shows will have results that come back that the father got only 96% or 98% proof of paternity, but still didn't question the results. (even if he didn't want to have to pay child support!)
In the 14-15 years since those tests were done, moreover, it has been shown that 20 to 30 to 40 percent of modern "Europeans" can be shown to have the same profile. pk
Yes the entire genome is vast, BUT every single human being shares the exact same sequences over 99.5% of the genome. The actual amount of variation is rather small, and occurs in what is called "junk DNA". Only certain strings of junk DNA will match with close blood relations. HOWEVER, only two or three max. mis matches will exclude relationship 100% no doubt.(she had five) THIS is why those claiming the mtDNA of Anna Anderson should be retested or is "unrealiable" are simply wrong.
To be clear. Whether one looks at 13 or 23 loci, ONE mismatch is an exclusion for mtDNA. The case of Anna Manahan had FIVE mismatches to the Victoria line of descent. There were NO mismatches for the Karl Maucher mtDNA. IF one examines 23 loci, the same five mis-matches will STILL be there. This is the reason that every single specialist in forensic mtDNA analysis says there is no reason to re test the Anna manahan samples.
For example (this is going to be extremely oversimplified, but ok for our purposes). Here are two sequences:
ACTGGGTAACGTAAGGTC
AGTAAGCCACTATACGCC
So we are comparing them to see if the two match. Normally you won't compare the entire sequence but just part of it...
So if we look at random loci (positions) then there is a chance we may get a false positive if we don't look at enough of them... Like this:
AC
TGG
GTA
ACGTA
AG
GT
C
AG
TAA
GCC
ACTAT
AC
GC
C
So even though these specific loci match, the sequence doesn't. Statistically speaking, the more loci you compare the more accurate the result will be, because in this case, if you did one more, there would be a mismatch and we would have our answer.
However, with mismatch, it's a different story once there is a mismatch, even of one base. So if you look at the same sequence but compare more loci and get one mismatch, it's a mismatch, period. There is no way you can a false mismatch... Like this:
AC
TGG
GTA
ACGTA
AG
GT
C
AG
TAA
GCC
ACTAT
AC
GC
C
And once there is a mismatch, all bets are off. ONE mismatch is all it takes, and she already has five, so it won't make any difference how many more sequences they test. One and she's out.