Is MyHeritage using a different map to compute cMs?

Question

It seems to me that MyHeritage is using a different map than other sites to compute lengths of DNA segments:

I have a triangulated segment with two other individuals.  MyHeritages's ChromosomeBrowser, says the shared segment is on Chromosome 12 from positions 4156627 to 16307468, and claims this segment is 22.6 cM.  

But this site:

http://compgen.rutgers.edu/map_interpolator.shtml

which I have used in the past to confirm length  calculations by gedmatch and FTDNA, tells me it should be more like 25.3 cM.  Am I missing something?

Comments

Your example is possibly within the range of variation between the different companies, but there are certainly some results that seem to me to be outright mistakes.

I imported an ftDNA result to MyHeritage, with one person I see a largest matching segment as 14.1cM on Chromosome 15 between 20004966 bp and 22852884 bp.  The Rutgers page say this is 6cM for females and 2.4cM for males, with a few tenths difference between build 36 and 37.3.

Not only that, but MyHeritage says there are only 128 SNPs in that range that make up the match. This region is next to the start of the hatched region on the chromosome browser, which I suppose is the SNP-poor region and whether it has something to do with this or with  comparing ftDNA with MyHeritage I don't know.

---

A number of people have reported a pile-up on that section of chromosome 15. That section should probably be redacted or get some exception handling in the code. I believe some have written to MyHeritage to suggest that.

Answer 1

Each company has its own algorithm, so you will see differences between each. Differences can, for expample, arise when one company might treat something as a single segment whereas others break it into two or more.