Go Back   Family Tree DNA Forums > Universal Lineage Testing (Autosomal DNA) > Family Finder Advanced Topics

Family Finder Advanced Topics Advanced discussion about Family Tree DNA's Family Finder Product.

Thread Tools Display Modes
Old 16th February 2018, 04:03 PM
Frederator Frederator is offline
FTDNA Customer
Join Date: Jul 2010
Posts: 815
Known issues with Gedmatch 'Lazarus' kits?

Just generated a Lazarus kit for my father from myself and two siblings, phased against our mother. I gave it a preliminary road test, and I have some questions.

Like why are known 4th cousins with segments >12 cM and 4,000 SNPs against the phased kits of at least two of the siblings failing to show even a 2 cM 100 SNP match against the Lazarus kit? Some of the no-shows are even >20 cM against the phased sibling kits.

I wasn't expecting perfection, but these are extremely well documented lines that seem to behave perfectly normally under what I had considered fairly good cross-examination. Are there any specific known issues with the Lazarus kits?
Reply With Quote
Old 16th February 2018, 04:17 PM
Frederator Frederator is offline
FTDNA Customer
Join Date: Jul 2010
Posts: 815
I don't think it should be an issue due to the specific location and robust quality of the no-shows I mentioned, but I have anticipated two questions that I'd like to put to rest right away:

1. I ran the diagnostic check against the Lazarus kit, and no specific issues detected.

2. Gedmatch determined that there were about 3,300 cM in the kit, but I think that's a bit misleading because, naturally, some of my father's paternal and maternal segments occupy the same coordinates on complementary chromosome copies.

I did a calculation in Excel for the total estimated # of cM in the Lazarus kit, both maternal and paternal, and came up with 5,969 cM. I think that's a pretty good calc. I based it on the volume shared between all of the siblings by one-to-one comparisons of their phased kits.

I think the formula is good because it returned a perfectly accurate answer when compared for the relatively simple trial case of two siblings (e.g., 3,587 cM inherited by each sibling from the father less 2,055 cM shared by the paternal phased kits for two siblings= 5,119 cM total paternal cM accounted for).

Just in case you were wondering, on average the siblings all seem to share a slightly higher than average volume of paternal data with one another. So on a pro-rata basis, for the # of siblings tested to date, we're around 310 cM short of our target.

But like I said, I don't think these aggregate statistics should have any impact on the relatively small (i.e., <40 cM), very localized no-show matches I've noticed to date. Just throwing it out there in case somebody knows something I hadn't considered.
Reply With Quote
Old 16th February 2018, 05:10 PM
Frederator Frederator is offline
FTDNA Customer
Join Date: Jul 2010
Posts: 815
Just to address another, maybe more serious question: I ran a map of the segments each sibling's UNphased kits share with the Lazarus kit in the 'Multi Kit Analysis' tool, and got some funky results. See the attached screen shot.

Now, I could understand if there were some significant gaps in the matching segments against Lazarus because SNP-poor regions aren't even tested in the first place. But those should apply to all brothers alike in a very uniform way, shouldn't they? Instead I got this weird, random-looking mess.

The siblings' kits were all purchased from FTDNA. Sibling I's was purchased maybe 5 years ago, but Sibling II was tested last year and Sibling III got their results just a couple days ago.

Is there maybe a problem going on with the phasing process of Lazarus creation? Comparisons of each siblings' kit to our mother's kit, and a direct comparison of the Lazarus kit to our mother's show no significant segments shared by our father and mother.
Attached Images
File Type: png Lazarus anomalies # 3 and # 4.png (48.6 KB, 4 views)
Reply With Quote
Old 16th February 2018, 05:48 PM
Frederator Frederator is offline
FTDNA Customer
Join Date: Jul 2010
Posts: 815
Originally Posted by Frederator View Post
. . .Gedmatch determined that there were about 3,300 cM in the kit. . . .
I guess this should have been a red flag. Less than 3,587 cM, which is about the amount usually returned by parent-child comparisons.

I guess I was so proud of my calc to gauge my progress against my goal of 100% cM capture that I didn't notice this fairly huge flaw.

Anyhow, that mapping exercise from the previous post should have made it clear that something isn't working as it should.
Reply With Quote
Old 17th February 2018, 09:13 AM
Frederator Frederator is offline
FTDNA Customer
Join Date: Jul 2010
Posts: 815
I see now that the phasing part of this process is indeed very sensitive, and I realize that I made a mistake in my original attempt. I have since deleted the original Lazarus kit and created a new, relatively more accurate Lazarus kit. Still far from perfect, but better than what I had before.
Reply With Quote


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Let's compare gedmatch kits MasterMel Recreation Room 1 22nd March 2017 05:52 AM
Has anyone used Lazarus in GEDMatch Lassie Recreation Room 2 24th July 2016 08:44 PM
FTDNA kits match on Gedmatch but not here? sfrank Family Finder Advanced Topics 11 26th February 2014 01:01 PM
One login multiple kits (Like Gedmatch) andbro Features Requests & Bug Reports Area 1 24th November 2013 02:48 PM
Gedmatch now has haplogroups for some kits Lincoln Family Finder Advanced Topics 10 2nd October 2013 08:25 PM

All times are GMT -5. The time now is 12:06 AM.

Family Tree DNA - World Headquarters

1445 North Loop West, Suite 820
Houston, Texas 77008, USA

Phone: (713) 868-1438 | Fax: (832) 201-7147
Copyright 2001-2010 Genealogy by Genetics, Ltd.
Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.