MyHeritage DNA full siblings only match about 38% - here is why

[melba_schmelba]

I only discovered this problem a couple of years ago, and found the explanation somewhere, which reassured me that my parents and siblings were in fact full siblings despite only matching around 38%. MyHeritage uses a different methodology to, I believe, Ancestry or 23andme, which involves not counting fully identical regions twice, this is explained well in this article I just found, written by Barry Starr in 2017, who I believe is now head of DNA at Ancestry

https://www.thetech.org/ask-a-geneticist/dna-shared-siblings

But it occured to me that this could actually be quite upsetting for people who did not understand or look into the DNA science, and as such I think it is rather irresponsible of MyHeritage to use this methodology since people could believe it means they do not share one parent

https://www.mumsnet.com/talk/_chat/4209679-My-heritage-dna-results-share-38-dna-with-my-sibling?page=1

[Gadget]

Thanks for sharing this, melba. 

Interesting, especially for the My Heritage methodolgy. It would explain the problems of  two of my cousins.


Gadget 

[brigidmac]

I think that's awful ..I imagine it would be very scarey for some people

My heritage should put out clear warnings that their Methodology is different

Melba + gadget did anyone on those lines retest on other sites. If so what percentage came out then

[melba_schmelba]

I think that's awful ..I imagine it would be very scarey for some people

My heritage should put out clear warnings that their Methodology is different

Melba + gadget did anyone on those lines retest on other sites. If so what percentage came out then

I agree, and I don't really understand why MH would choose to do it this way. All the kits I mentioned are uploads, and they show as 49-51% sharing on Ancestry and 23andme.