The only cousin statistics that acknowledge the differences in paternal and maternal relatives due to recombination rates.
See the relationship probability calculator that’s based on this same model.
I’ve written an article that shows the percentages of shared DNA that a person shares with various relatives and ancestors. These are the only tables or charts of shared autosomal DNA (atDNA) that can be verified by standard deviations from peer-reviewed studies. Other charts or tables have much lower standard deviations, probably because of removing data erroneously thought to be outliers.
Since many in the genetic genealogy community prefer to use centiMorgans (cM) over percentages, and since cM aren’t the same from one platform to another, I’m going to display cM values for each platform on separate pages. The tables here are only for AncestryDNA kits. Many of the values shown here include fully-identical regions (FIR), which can’t be found at AncestryDNA. However, these values could be seen at GEDmatch if an Ancestry user has uploaded there and is comparing to another kit from AncestryDNA. And I do recommend using the total cM for the platform from which you uploaded when converting percentages to cM unless one has made a superkit at GEDmatch.
Table 1. Shared cM of atDNA between siblings for AncestryDNA kits. HIR = ‘half-identical regions,’ where one of the two chromosome homologues matches. FIR = ‘fully-identical regions,’ where both copies of a chromosome match. HIR + FIR = all of the points on chromosomes where two people match once plus all of the points where they match on both copies. HIR counting includes FIR bp, but only counts them as if they’re half-identical.
Table 2. Shared cM between six different types of 3/4 siblings for AncestryDNA kits. All parameters are the same as for Table 1.
Table 3. Shared cM for double first cousins with AncestryDNA kits. All parameters are the same as for Tables 1-2.
Table 4. Shared cM for grandparents and some of their descendants using AncestryDNA total cM. All parameters are the same as for Tables 1-3.
I hope you’ve found these results useful. More will be on the way.
Feel free to ask me about modeling & simulation, genetic genealogy, or genealogical research. And make sure to check out these ranges of shared DNA percentages or shared centiMorgans, which are the only published values that match peer-reviewed standard deviations. That model was also used to make a very accurate relationship prediction tool. Or, try a calculator that lets you find the amount of an ancestor’s DNA you have when combining multiple kits. I also have some older articles that are only on Medium.
This data is very useful as I have a statistics background. My question is, however. Where do I find centiMorgan data for the general population and up to 5th through 8th cousins? I want to run some hypothesis tests. Also is there a database that will give you raw shared cM all the way down to zero regardless of whether they it is considered a “match”
Hi Douglas,
I have cM ranges for pretty much every relationship to 15th cousins, but I haven’t gotten them online yet. I’ve never seen any demand for that before. I’ll try to get that online, but I always have several projects in the queue. I don’t know of any dataset that includes matching people as well as randomly chosen non-matching people from the population.
I apologize, but I am confused. I don’t have a great grasp on statistical data. Can you please explain this? Subject A and B are first cousins. I share 118 cM’s with A and match as a predicted third cousin. However with B I only share 21 cM’s and match as a predicted 4th cousin. That seems like a large range. Is that normal?
Hi Sarah,
This article answers the question of how differently two siblings can match a cousin: https://dna-sci.com/2021/07/19/why-does-my-sibling-match-a-3rd-cousin-and-i-dont/
question please: Ancestry tells me that I share 3,450 cms with my mother (mother/daughter) and 36 segments, longest 282. I would have expected 23 segments — How does Ancestry come to the count of 36? I find this very number very problematic. Thank you, J. Ross
Hello,
This is 100% due to genotyping errors. You actually share 23 segments with your mother, all of which are full chromosomes. Ancestry only counts autosomal DNA, so you share 22 segments over 22 chromosomes with your mother. But Ancestry is notorious for egregious genotyping errors. It may be that all sites have just as many errors in their kits, but every other company seems to be much better at recognizing those and fixing them with their matching algorithms. At Ancestry you might see your actual number of segments quadrupled when they report it.
Hello my brother and I did ancestry dna kits and we matched at 1795 cMs and 56 segments. We always thought we were full siblings but doesn’t this indicate that we are half siblings?
Hi Kathi,
Yes, sorry, that’s too low for full-siblings. SegcM gives you a 50% chance of being maternal half-siblings and a 0% chance of being paternal half-siblings: https://dna-sci.com/tools/segcm/
Hello! Your tables and information have been very helpful.
I am trying to solve an old family mystery. My Gran’s father had an affair with his wife’s sister, so we believe some of the aunt’s children shared a father with Gran. So, some of her double cousins are actually 3/4 siblings. Unfortunately, gran is 97 years old and most of the cousins are dead and unavailable for testing. Some of their children and grandchildren have tested and matched with Gran on AncestryDNA. Is there a formula to create numbers for shared % of DNA or cM shared for the children and grandchildren of a double cousin and 3/4 sibling so that I can better determine who fathered the children?
Hi Ruthie,
I’m glad you’ve found the tables helpful! There’s no mathematical way to calculate ranges of shared DNA. But the averages can be calculated simply by adding the averages of each relationship: https://dna-sci.com/2021/01/05/can-you-just-add-the-averages-and-ranges-for-double-relationships/
If you use percentages to calculate the totals, you can easily convert them to cMs for each site, although there’s nothing wrong with percentages. For Ancestry, you can multiply percentages by 69.78. For GEDmatch, it’s 71.75. For two males at 23andMe, it’s 72.58. For anyone else at 23andMe, it’s 74.4.