Announcement

Collapse
No announcement yet.

Determination of centiMorgan Values

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Determination of centiMorgan Values

    This is a two-parter. First, the FTDNA Learning Center states that the Family Tree DNA bioinformatics team works with centiMorgan (cM) data from the International HapMap Project to evaluate genetic distance/recombination rates. However, NCBI retired the HapMap Project in June 2016, and is evidently deferring to the 1000 Genomes Project. NCBI does, though, state that archival HapMap data will continue to be available via FTP for an unspecified period of time.

    Is FTDNA continuing to use data from the retired HapMap Project, or have they transitioned to a different mapping resource?

    Second, most common linear interpolation methods used (e.g., Kosambi) to estimate cM by base pair location vary not only by the human genome reference assembly build in use (e.g., Genome Reference Consortium hg36 NCBI Build 36.1; GRCh37 Build 37.3) but also by gender, with cM values often differing significantly for the same bp range if mapped against the female vs. male genome.

    Does FTDNA exclusively use "sex-averaged" centiMorgan estimates, or does it attempt to reflect the gender-specific mapping for each Family Finder kit?

  • #2
    As far as I know, all the vendors use the average of the male and female recombination data for the purposes of the standardized human genetic map expressed in centiMorgans. This makes sense, otherwise we would have to jump through several additional hoops in order to interpret relationships based on the lengths of matching segments. However, regardless of the standardized segment lengths, it may sometimes be useful to consider the differences in terms of the probability of obtaining certain results. For example, the probability that a certain segment comes down intact through 5 generations, all male, should certainly be greater than the probability of the same segment coming down through 5 generations of females.

    Comment


    • #3
      Originally posted by John McCoy View Post
      . . . For example, the probability that a certain segment comes down intact through 5 generations, all male, should certainly be greater than the probability of the same segment coming down through 5 generations of females.
      I have spent an unreasonable amount of time and effort trying to quantify these probabilities. But at the end of the day I have to say that the increased volatility in male inheritance effectively offsets the true fact that you reference, namely that the male recombination rate is lower than the female rate. You're trading off greater odds of inheriting an intact segment whole against the greater odds that it will fall within the range of segments eliminated through recombination. A complete wash.

      Comment


      • #4
        Originally posted by Frederator View Post
        I have spent an unreasonable amount of time and effort trying to quantify these probabilities. But at the end of the day I have to say that the increased volatility in male inheritance effectively offsets the true fact that you reference, namely that the male recombination rate is lower than the female rate. You're trading off greater odds of inheriting an intact segment whole against the greater odds that it will fall within the range of segments eliminated through recombination. A complete wash.
        When performing a statistical analysis, the first step is to arrive at an estimate of the population average value. The next step is to look at how consistently individual observations approximate the average value (i.e., standard deviation). In this case it is my experience that the higher estimated population average survival rate is offset by the greater variation in observed values to the point where there is effectively no difference in the survival of male vs. female ancestral contribution. This was a very disappointing finding, but at least I am satisfied that I now know for a fact, rather than merely speculating without any objectively measurable basis.
        Last edited by Frederator; 24th January 2018, 05:32 PM.

        Comment

        Working...
        X