When doing triangulation, I can end up with a segment shared among 3 in a GEDmatch compare where I have the cM for each pair of the three, but not the cM for the overlapping portion of the segment. I was unable to find a calculator where I could input the chromosome number and the beginning and ending position numbers. For example, I'd like to know the cM for a segment from 217,656,670 to 227,193,370 on Chromosome 1.

up vote 10 down vote accepted

There is a service to turn a physical chromosome position (measured in base pairs, "bp") into a linkage position (measured in centiMorgans, "cM") here:

To get the cM for a segment given its begin and end positions, you would use the web site to convert each of the two positions into linkage positions and then subtract the start linkage from the end linkage.

enter image description here

For your chromosome 1 segment from 217,656,670 to 227,193,370 you should get linkage position from 231.07 cM to 237.65 cM. The length is then (237.65 - 231.07) = 6.6 cM.

enter image description here

Notes on using the Rutgers site: You should put one position per line into the input box, and use only digits, e.g. just "217656670". You choose the build (probably 36) and the chromosome number from the dropdown lists. Under "Query Header" choose "Absent". On the results page, you probably want to use the value from the "Sex-averaged (Kos cM)" column. Males and females have quite different linkage values, but genealogical segments are generally characterized by the average of the two values.

  • Robert: I was writing up the same answer, only to see you post 3 minutes before me, so I deleted my answer and added my two useful photos to yours. – lkessler May 19 '17 at 5:16
  • LKessler, thanks for adding your screenshots into the answer; it definitely improves it. – RobertShaw May 19 '17 at 5:18
  • Seems like there needs to be a caveat on this tool. I just compared the Rutger calculations to what FamilyTreeDNA gives in its Chromosome Browser Results file. They are different. e.g. A FamilyTreeDNA match between a male and a female on Chr 1, 1794162 to 3789978 lists at 5.16 cM, but Rutgers gives 5.43 female, 8.39 male, and 6.57 sex-averaged. FTDNA got their mapping from the International HapMap project, whose numbers obviously are different than Rutgers. See: familytreedna.com/learn/autosomal-ancestry/… – lkessler May 19 '17 at 5:51
  • It seems that Rutgers does match GEDmatch, but not FTDNA. – lkessler Aug 4 at 23:05

Hendrik Wendland has created a free tool for Windows (donations accepted) called MapS Converter that you can download. It can calculate genetic distances and convert between Build 36 and 37. It is available at Hendrik's MapS Phasing site: www.maps-phasing.com

enter image description here

For example, in the screenshot, I entered the Base Pair addresses you gave as an example in your question into the Build 36 values, I then press the two arrows at the bottom right. It then calculates the Build 37 values and the genetic distance of 6.58 cM.

The value is sex-averaged. By selecting two women or two men at the bottom, you will get female and male values.

Alternatively, you can enter Build 37 numbers (if your numbers are from 23andMe or Living DNA) and get Build 36 equivalents. Or you can enter a start location with a cM value and it will return the end location.

It is a nice little tool, well designed, and easy to use.

I didn't see such type of calculator. But I can imagine the next algorithm:

  1. As we know each chromosome has it's own value for number of base pairs and cM. Also there is no direct connection between bp and cM, because bp is physical number of pairs and cM is the measure of recombination probability (or rate). Please note that we don't know HOW MUCH base pairs are precisely in your chromosomes, but the human reference genome will give us enough precision despite it can differ from yours particular.
  2. So we can can took table data for whole length choromosome in bp and cM. The table data can be taken from here. Also wikipedia has info about number of bp in each chromosome.
  3. And then use linear approximation:

your_cM = (whole_chr_cM) / (whole_chr_bp) * (end_bp - start_bp)

  • P.S. if somebody has more interesting idea about calculation, I will be glad to see it – GEORG GAAL May 18 '17 at 5:52
  • 1
    This approach is likely to yield an inaccurate result since the linkage (recombination rate) varies substantially in different places on a chromosome. – RobertShaw May 19 '17 at 4:41
  • @RobertShaw I totally agree with you that recombination rate is not the same on the whole length of chromosome. But it is the simplest and most straight-forward method available to everybody. – GEORG GAAL May 19 '17 at 6:52

Your Answer

 

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.