# what are average, median, and minimal dna closeness of stranger peoples, compared to closeness of relatives?

i thought: a parent closeness to a person can be marked as 50%. seems that is 50% of all parts of dna of humans that have variability across all human population. i heard that chimpanzees are 98% close to humans by their dna, so, 100% here may match like absolute 1 or 2 % of dna.

then, if to say only about siblings with both parents same, for simplicity, a sibling also must have 50% closeness in average. then, an uncle has 50% * 50% = 25% . (because he is sibling of a parent). a cousin has 12,5%. (child of uncle). a grandfather is 25%, his sibling is 12,5%, son of his sibling is 6,25%, and grandson is 3,125%, that is, second cousin. third cousin should be 0,78125% .

and here a question appears: so, this number goes very low very quickly, when going to farther relatives. but it should not go to 0, because stranger people also must have some similar genes by chance.

so, i want to ask, what is that average level of the similar genes by chance? what is median level of it? and what is the distance between 2 of the most far persons?

i think, maybe this question is not possible to answer nowadays. because i searched, browsed in wikipedia, and saw that, seems, it is not known, to how many parts chromatids are split in meiosis. so, must be, it is not known, how much is that variability of human dna. i think, maybe this question may stay waiting for future answers.

• this seems to give info for that: google.com/… would the fourth image on this page give the information that you wish for? quora.com/… Commented Jul 20, 2023 at 9:08
• Everyone is related, it just gets harder to know exactly how related you are when you go beyond a handful of generations. When you talk about "shared" DNA with close family members, you're talking about shared variation. When you talk about "shared" DNA between chimps and humans, it's more like you're talking about similarity of the average. These are two very different, not-comparable measures. Commented Jul 20, 2023 at 13:35
• The 1-2% difference refers to SNPs in coding regions. There are drastic higher-order differences like chromosome fusion (chromosome 2a/b) and deletion or reversion of long stretches in MB level. What you are calculating is theoretical inheritance of alleles by identity of person from which it comes, which is abstract and has nothing to do with the actual sequences. Commented Jul 20, 2023 at 15:10
• I am speaking of chimpanzee vs human. Commented Jul 20, 2023 at 15:10

## 0.1% and 0.

the average person varies by around 0.1% from another random person, or two random people are 99.9% the same. Humans actually have very low genetic diversity as a species. Total variation is around 0.4% by copy number. Of course the catch is it all depends on how you calculate difference. Different ways of calculating it will give you slightly, but not insignificantly, different answers. For instance you have issues with duplication and deletion, humans can have multiple copies of the X chromosome with little effect, is a translocation one difference or do you count each difference in a 1:1 base pair map, do you count genes if we don't know if they code for anything, ect. There is a whole rabbit hole of competing ways to calculate genetic difference.

the 50% you are referring to for relatives is really 50% of the parts that vary, or 50% of 0.1% thus 0.05%, or you are roughly 99.95% identical to each parent and 99.6% identical to the most unrelated person on the planet, again depending on how you calculate it.

The minimal difference is 0, identical twins who don't know each other exist. Also the cutoff for relative is completely arbitrary, there are humans who don't share 100% of genes within themselves.

Mean in this case does not make a lot of sense, the closest you might get is something like a genetic distance map, because it is not a single variable axis. there is a whole wiki devoted to human genetic variation if you want to dive deeper. https://en.wikipedia.org/wiki/Human_genetic_variation

More sources

More on how relatedness works.