June 20, 2013

How Much DNA Do You Share With Your Blood Relatives?

How much genetic material do you share with your biological relatives?  Take a look at the family tree below and the figure in the red box is the percentage of DNA your body has in common with your blood relatives.  For example, your first cousin has 12.5% of the genes you do (implying, inversely, that 87.5% of their genes are different).  Your third cousin twice removed, on the other hand, would have only 0.195% of the same genes, meaning 99.805% of their DNA would be different.  This assumes, of course, that you have no double relations in your family tree (e.g., sharing a great-great grandmother from two sides of your family tree).  

DNA Shared Between RelativesThe chart also ignores relatively rare phenomenon such as the elusive double cousin.  These relationships arise when two siblings of one family reproduce with two siblings of another family.  This results in the children being related to each other through both parents, and sharing the same grandparents.  As a result, double cousins are genetically equal to half-siblings, sharing double the DNA normally seen in first cousins.  

  • Doug

    Not sure I understand. I would expect to share well over 99% of DNA with my immediate relatives. How do you calculate that I share only 50% DNA with them?

    • Joshua Kennon

      Your body was created by taking roughly 50% of the genes from your father and 50% of your genes from your mother.  If you have siblings, different genes in different combinations were passed onto then due to sexual reproduction creating a roulette effect, resulting in siblings sharing roughly 50% of the genetic material, as well.  

      The exception is identical twins, who share 100% of DNA due to coming from the same egg and the same sperm, which split and began reproducing independently from the same gene pool combination.

      On a related note, a common misconception is that identical DNA must result in identical outcomes.  Identical twins, for example, have different fingerprints, which you can read about here: http://www.nytimes.com/2009/10/06/science/06qna.html.  If one twin suffers from autism, there is only a 90% chance the other will, too (not 100%).  If one twin is gay or lesbian, there is only a 50% chance the other is too (not 100%), etc. etc.  This is because of, in simplified terms, the fact that our genes have ‘on/off’ switches attached to them that can be flipped through conditions in the womb, environment throughout life, exposure to certain chemicals, and a host of other factors.  For a very basic explanation of how genes get flipped on or off, read http://www.thetech.org/genetics/ask.php?id=63

      And, of course, it’s fascinating how our ancestral DNA plays a role in our day-to-day lives.  Most of the humans alive on planet Earth cannot drink milk; the figure is around 70% for adults.  If you live in the United States and are white, though, you wouldn’t realize this because your ancestors developed a mutation that allows us to process cow’s milk.  This explains more: http://www.thetech.org/genetics/ask.php?id=250

      If you prefer a hierarchy written out, Genealogy.com has that at http://freepages.genealogy.rootsweb.ancestry.com/~laetoli/degree.html

      • Doug

        Ahh ok.. it is just a semantic difference then. I was comparing your percentage to statements like “Humans share 98% of their DNA with chimps”. That percentage is about how much of the data contained in the DNA is the same regardless of ancestry, where you are talking more about the bloodline path of where it came from.  Thanks for the clarification!

        • http://andrewdouglass.com/ Andrew Wells Douglass

          I agree the “DNA shared” figures are misleading. They make it sound as if there were an infinite numbers of genes, when of course there are only so many ways to code for whatever protein. Perhaps it would be better stated as “probability you share a given gene because of shard ancestry” for, say, eye color (which is itself polygenic, not blue-or-brown as taught in grade school!). Some genes also tend to be transmitted together. You share some/many genes by pure coincidence. Thanks.

      • TL Howard

        Actually, the most recent and large sampled research (Australian Twin Registry) plus another (forgot its name, sorry) shows that the concordance of monozygotic for homosexuality (male) is only 20%. There is obviously a biological cause/trigger (with only a genetic susceptibility to that environmental trigger) to male homosex.

  • K

    What are the figures for half-sibling relationships?

  • Tashina

    I found out I have a cousin whos dad is my mom’s scound cousin. And his mom is my dad’s scound cousins aswell, how much DNA do we share?

  • http://www.joshuakennon.com/ Joshua Kennon

    This is a joke, right? Please tell me you realize the numbers shouldn’t add up to 100% …

  • http://www.joshuakennon.com/ Joshua Kennon

    Again, this is a joke, right? …

  • John Tate

    Why would the numbers add up to 100

  • jjay09

    why is it that even charts displaying the family tree is done so in a such a sexist way

    • http://www.joshuakennon.com/ Joshua Kennon

      Given that there were two possible ways this chart could have been presented by whomever designed it, it is impossible to say whether or not the chart is presented in a sexist way as there was a 50% probability of either the paternal or maternal lineage being used as the illustration (a complete family tree would have been more difficult to read and therefore not as optimal to utilize as a visual aid so we will exclude it from the analysis).

      Absent other information, a determination of sexism is impossible. For example, had a maternal family tree been used, a man could have claimed it was sexist but it would not make it so. There simply isn’t enough evidence to indicate either way.

      Without said evidence, or any additional supporting data, the conclusion that the chart is sexist is a form of subconscious confirmation bias meant to strengthen a present conviction in the observer that reinforces preconceived ideas.

      (An an interesting discussion of second and third order effects, one could argue, upon studying, say, 100 textbooks with family trees in them, that a distribution that represented more than an approximate 50/50 split would be sexist but even this would be problematic as people are influenced by their life circumstances. In the lower and middle classes, nearly 1 out of 2 boys are now born to a mother out of wedlock, which would theoretically lead to a higher probability of them using her bloodline if they were preparing a chart such as this, skewing the results in favor of women. On the other hand, given past legal precedence based on the nearly now-defunct patriarchal structure of marriage a century ago, names are passed down from the male in almost all cases as a result of cultural practice, meaning that a person might be more interested in knowing where his or her namesake originated due to a myriad of subconscious forces, such as mere association tendency (which is so powerful, there is overwhelming statistical evidence that boys with names like “Dennis” become dentists far more likely than should be the case; it’s the way the human mind pulls us in certain directions). Therefore, you could expect the distribution to favor males. Still yet, whomever prepared the chart could be a medical student studying Y-STR on the Y-Chromosome and want to illustrate the unique situation found exclusively in males that causes each and every person on the male only chart to share an identical copy of nearly half the genetic material that composes their body.)

      TL;DR: It is mathematically impossible based on present evidence to conclude whether there was any form of gender preference in the creation of the chart.

  • stacie

    I have a first cousin that i believe is my half brother what would the percentage be if we did a dna test?