Not a Real Match? An Update Using 4 Generation Phasing

updated 11 Sept 2015 to include maternal matches

You Can’t Inherit Anything from Ancestors that your Parents didn’t Give You

That seems pretty straightforward, doesn’t it?  Traits can skip generations and not be expressed, even though they may be there, hiding, in your DNA. The DNA itself, however, cannot skip generations. If your parents didn’t have it, neither can you.

When I began my DNA adventure trying to figure out who I inherited my matches from was half the fun.  Unfortunately, I knew they were only guesses and for quite a few of my matches I had no guesses at all.

My dad tested next, and when his results were in, I was able to divide my matches into 3 piles: anyone that also matched him was paternal; anyone who matched me at 10 cM or above but did not match him was maternal; anyone who only matched me between 7-10cM was tentatively considered a maternal match*.

I manage several accounts and use Genome Mate to keep track of each person’s matches. In Genome Mate, not only can I look at each profile’s matches chromosome by chromosome, but I can call up additional information by clicking any match’s name.  Rather than having to jump from profile to profile to see which matches were inherited from my paternal grandmother or only my dad, the Profile Overlaps box gives me that information at a glance:

Featured image

Karen: Figure 1

In this example on chromosome 9, my grandmother G Duffy, my dad, myself and my sons all match Karen at 13 cM.  Genome Mate allows me to apply a feature, the In Common With  (ICW) label to organize our matches.  In this same example there is a ? beside my grandmother as I don’t yet know whether she inherited that segment from her mother or father.  For my dad it is a M (maternal) match, for myself it is PGM (paternal grandmother) match and for my sons it is MGF (maternal grandfather) match.  With each match labelled, I know which quadrant of the profile’s tree to look in, no matter whose profile I am handling, to narrow down the search for a common ancestor.

Why Aren’t We Matching?

When my paternal grandmother’s results came in I was able to divide our piles even further and that was when I noticed a problem. Some of the matches I’d attributed to my maternal side, as they didn’t match my dad, matched my paternal grandmother.  When I checked my dad’s account at FTDNA, I realized it wasn’t an oversight. The matches my grandmother and I shared, and often one or both of my sons, did not appear on his Family Finder Match list.

Featured image

Howard: Figure 2

Had there been a problem with his sample?  Was it a result of no calls?  FTDNA won’t allow a match to appear unless it meets minimum requirements, and all of the matches were imported directly from our respective match lists at FTDNA.

Which Matches were Genuine?

Some DNA companies provide phasing, but FTDNA does not.  Since you can’t inherit anything your parents didn’t give you, some matches needed closer scrutiny and begged the question, “Which matches were genuine?”

The shared segment between Howard, my grandmother, my sons and I in Figure 2 is on chromosome 9  and is 9.5 cm.  Pay close attention to the start and stop points:

  • Grandmother Gladys 36-74
  • Self Lori 36-74
  • Son Jeremy 36-74
  • Son Gavan 36-76

Notice they are very similar to the start and stop points also on chromosome 9 with Karen, from Figure 1:

  • Grandmother Gladys: 33-74
  • Dad:32-74
  • Self Lori 33-74
  • Son Jeremy 32-74
  • Son Gavan 33-76

Using FTDNA’s chromosome browser, the dark colored base figure represents chromosome 9 for Gladys. The match between Gladys and Howard is represented in orange (9.5 cM at FTDNA 10.4 cM in Genome Mate) and the match with Karen (13 cM) is represented in blue.


Figure 3 Shared segments between Gladys, Howard and Karen

Gladys, Howard and Karen appear to share an ancestor.   When looking at each match in isolation, the conclusions seem straightforward:

Manual Phasing for our match with Howard:

  • Grandmother Gladys 36-74: ? Paternal or Maternal match
  • Dad: no match at FTDNA
  • Self Lori 36-74
  • Son Jeremy 36-74
  • Son Gavan 36-76

Manual Phasing for our match with Karen:

  • Grandmother Gladys: 33-74 ? Paternal or Maternal match
  • Dad:32-74 Maternal match, although the start point was paternal DNA and not inherited from Gladys
  • Self Lori 33-74 Paternal Grandmother
  • Son Jeremy 32-74 Maternal Grandfather
  • Son Gavan 33-76 Maternal Grandfather

What’s the Problem?

For Gladys, Howard and Karen to share an ancestor, my dad can’t match with one and not the other.  Since he matches Karen for the entire 13 cM segment she shares with Gladys, but does not match Howard, it appears that Karen and Gladys are genuine matches.

Despite the size of the matching segment, Howard can be considered either a false match, or for my dad a genuine match not meeting the minimum 20 cM requirement at FTDNA.  To be certain, we would need to compare our kits at Gedmatch.  To this date, every match at FTDNA who has uploaded their kit to Gedmatch has been a match for the correct generations (example: grandmother/father/self/child) even if they are missing from the match chain at FTDNA.


Figure 4 Chromosome 9 for Gladys showing shared segments with Howard, Karen and her son Steve.

As you can see, this wasn’t an isolated event:

Featured image

Figure 3

The shared segment between my grandmother and I in Figure 3 is 8.3 cm.

Featured image

Figure 4

The segment in Figure 4 is 8.1 cm and is shared between my grandmother, son and I.

Featured image

Figure 5

The shared segment in Figure 5 is 8.6 cm and is shared with both my sons, my grandmother and I.

Maternal Phasing

This phenomena isn’t limited to my paternal side but it is seen much less frequently in my maternal grandmother’s match list .  There are two reasons for this.  My mother has not tested which limits our phasing ability. Instead, my maternal grandmother and 2 daughters of my maternal grandfather have tested.  Also, my maternal grandmother has very few matches compared to my Irish paternal grandmother, as they have very different ethnic backgrounds.


Leona Figure 6

In Figure 6, my grandmother Leona is compared against her daughter Jackie, myself and my two sons.  Jeremy’s stop point of 89 matches Leona’s DNA, but was not inherited from me.

A similar situation occurs in Figure 7 where my son Gavan inherits his start point from his father rather than me or his great grandmother Leona:


Leona Figure 7

With the two daughters of my maternal grandfather added to the mix, the number of false matches rose.  Since Jackie and Diane are half sisters, and both daughters of Jim, and Jackie’s mother Leona has also tested, the group has been invaluable for weeding out false matches.  Again, for certainty, matches must be checked in a 1 to 1 comparison at Gedmatch rather than relying on FTDNA.

Chromosome 12 was chosen at random, and the process produces similar results across the board.


Lori’s Maternal Chromosome 12

For Chromosome 12, all paternal matches have been removed. As she has tested, we automatically know which segments are inherited from Maternal Grandmother (MGM) Leona.  What isn’t inherited from her is likely inherited from Maternal Grandfather (MGF) Jim.  Since 2 of Jim’s daughters, sisters of my mother have tested, anything that matches them but not Leona can be considered a genuine MGF match.  Anything that doesn’t match Jackie or Diane, may have been inherited from my mother, is a false match, or is potentially a match not meeting the minimum FTDNA match requirement but will prove a genuine match at Gedmatch..

To determine whether a match is valid, my DNA is compared against start and stop points of known relatives:

  • Jackie 0-19 (not shared with her mother Leona=inherited from her father Jim)
  • Diane 74-102 (not Leona’s daughter=inherited from  her father Jim)
  • Jackie 102-132 (shared with her mother Leona=not inherited from her father Jim)

On Chromosome 12 all the matches except for 2 cluster neatly into groups that clearly match Jackie and Diane or Leona.  That leaves Camilla and Patricia needing closer examination.

Camilla: 48-60 10.8 cM.  She doesn’t match Leona.   Neither Jackie nor Diane and I share the 48-60 segment.  The match may be genuine, and inherited from my mother from her father Jim.

Patricia: 118-123 8.1 cM.  She doesn’t match Leona.  Jackie and I share the segment 102-132, known to be inherited from Jim, but Patricia doesn’t match Jackie. Either Patricia is a false match, or she and Jackie don’t meet the minimum 20 cM requirement at FTDNA.

With 4 generations and two pairs of half siblings tested, it becomes clear that phasing is an extremely valuable tool. Generally, I prefer to work with matches 10 cM or more, and my experiment with phasing has reinforced that idea.  However, phasing has also proven useful in confirming matches below 10cM (I have only focused on matches above 7 cM), some of which have held on through all 4 generations.

*More information about Identical by Descent inheritance can be found here.

FTDNA defines IBD as, “IBD stands for Identical By Descent. This means the DNA matches because it comes from a common ancestor. IBD can refer to a single mutation or to a segment of DNA. If a mutation or segment of DNA is IBD among a group of people, it comes from a common ancestor.

The Family Finder relationship predictions require a minimum number of results in a row to be identical in order to identify that the segment is likely to be IBD.”


IBS as, “IBS stands for Identical By State, meaning the DNA matches by coincidence. When two individuals share numerous individual results without being related, those results are IBS.”


3 thoughts on “Not a Real Match? An Update Using 4 Generation Phasing

  1. Hello
    I manage the Clan Cameron DNA web project and we have similar problems with missing data between generations that can’t be explained.
    Have you received any information from FTDNA?
    Thanks Kim


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: