Affy to Illumina conversions complete?

[JohnG]

Hold tight

Matt indicated there is some review in process, so best to hold tight for now.

I have saved my Affy match lists so I can still email folks and I intend to work on the connections with some of them.

It does not look to me like a SNP difference. Some of my Affy matches still match my immediate family, just not me. I have lost some marginal matches and some very substantial matches and some matches with paper trails that prove them. I think it is a process thing to be fixed.

[GayeSherman]

Before and After

Affy: 254
Illumina: high of 326, now 257 (post Jewish recalculation)

Missing: 113 Affy (75 never in Illumina, the balance were there but now gone)

Among the missing: 2 Third Cousins and 11 Fourth Cousins

[cbon20]

Total Affy 106
Total Illumina 94
matches common to both Affy and Illumina 51

[mkdexter]

News!

Thanks for the answers. I received enough at this time and I've sent the data on incase they didn't read it here.

Yes they are looking at this. I talked to the office manager, Jane, about an hour ago and this is what I was told.

1. I.T. came back and says there are NO software bugs. There is a bug in yDNA reporting but that's different, not related to Family Finder, and its been announced to project admins.
2. It will take some time to review this issue but they have taken some data and are going to look into it farther.
3. They probably will not take down the Affy lists next Tuesday, but a week later.
4. They don't think there is a large difference the algorithms causing this, just that the SNPS tested are different.
5. There are 100 tests still in process of conversion, just 100 not more.

This is something they need to speak to so I'm not planning on talking to them anymore for a few days, plus they'll be busy anyway.

Max is heading up everything on this issue now.

I believe I know what they will say and in my opinion it will be something about the tests being different and that Affy matches misisng do not mean they are invalid, just they have such small quantities of DNA they were not detected by Illumina, a test with more data points.

I reiterated we the customers need to see something "official" from FTDNA and that with the 4 lists I sent (thank you JohnG) they need to rake a few kits and review the raw data and determine why the Illumina match was not detected while the Affy was.

They are looking and they will need some time but no softare bugs found.

[dwight]

Thanks for the liaison work and update, Matt. Much appreciated!

Thanks for the answers. I received enough at this time and I've sent the data on incase they didn't read it here.

Yes they are looking at this. I talked to the office manager, Jane, about an hour ago and this is what I was told.

1. I.T. came back and says there are NO software bugs. There is a bug in yDNA reporting but that's different, not related to Family Finder, and its been announced to project admins.
2. It will take some time to review this issue but they have taken some data and are going to look into it farther.
3. They probably will not take down the Affy lists next Tuesday, but a week later.
4. They don't think there is a large difference the algorithms causing this, just that the SNPS tested are different.
5. There are 100 tests still in process of conversion, just 100 not more.

This is something they need to speak to so I'm not planning on talking to them anymore for a few days, plus they'll be busy anyway.

Max is heading up everything on this issue now.

I believe I know what they will say and in my opinion it will be something about the tests being different and that Affy matches misisng do not mean they are invalid, just they have such small quantities of DNA they were not detected by Illumina, a test with more data points.

I reiterated we the customers need to see something "official" from FTDNA and that with the 4 lists I sent (thank you JohnG) they need to rake a few kits and review the raw data and determine why the Illumina match was not detected while the Affy was.

They are looking and they will need some time but no softare bugs found.

[rod]

There are 100 tests still in process of conversion, just 100 not more.

Does that include the people who had to submit new samples?

[LSSmith]

How many Affy matches are people now missing?

Missing Affy matches compared to original number of Affy matching kits:

Myself: 33 out of 73
my mother: 33 out of an 66

Only one person in both my missing and my mother's missing is the same match to us both..

I also have a few kits that showed up for either myself or my mother but not both whereas before they were to us both.

Mother:
25 Affy.......1 converted to illumina
Total of 7 matches.....6 new ones.

Me:
49 Affy.......10 converted to illumina
Total 21 matches....11 new ones

My mother has lost about 95% of her Affy matches and now has less than 1/3 total matches than before the conversion. Can anyone beat that?

[mkdexter]

Does that include the people who had to submit new samples?

I don't know rod.

I emailed these questions this morning and then followed up with a phone call to the office a few hours later and spoke to Kathleen and Jane.

Questions:

1. At this point it sounds like almost everyone has lost around 50% FF speculative Affy matches. Is there really a software bug?
2. Also a few people have said they did not get their converted kit results back yet - so how can all the conversions be complete?

____
On the second question Jane said there are 100 kits pending which they thought would be done by Thursday but are still out and should be done soon. I don't know what they are other than that there are 100 still out.

[swimmingly]

I just counted, and 14 out of the 41 total names on my Affy list carried over to Illumina. I think it is due to the different sampling.

[JimH]

I just counted as well, and out of 118 Affy matches, I lost 54. One was a predicted 4th cousin, and 53 are from the speculative category. I hope FTDNA can explain what we're all seeing.

[royfarnol]

My score is:-

Affymetrix 55 matches, 6 at 4th cousin.

18 have resurfaced in Illumina together with 28 new matches with 5 at 4th cousin.

My 37 lost cousins - where are you?

[JPHutchins]

65 Affy matches.
25 Affy conversions to Illumina.
32 Illumina only matches

That leaves 40 Affy matches that are missing. Most are 4th or 5th to distant.

[goshelli]

Affy to Illumina

43 Affy, 20 converted to Illumina, 23 MISSING.

My brother picked up one of my missing Affys.

21 of my missing Affys were listed as 5th to distant cousins.

I'd like to think I should only consider a match a match if we share a 10 cM segment. However, the ONLY DNA/paper trail confirmed connection I have made is with a 4th cousin, once removed. Our largest shared segment on Affy is 8.49 cM and 8.29 on Illumina.

[nolnacsj]

Affy matches: 75
Affy conversions that are Illumina matches: 41
Affy matches not showing up in Illumina: 34

Thanks!

Kit N68639

Original Affy Matches= 33
Affy conversions now showing up in Illumina= 7
Affy matches not showing up in Illumina=26
Total present Affy/Illumina matches=26
Date last Affy also showed up as a Illumina Match- 4/19/11
Date I received last new Illumina Match-4/15/11

Date last new match(s) was posted to my Affy was back on 1/14/11 (2).

I am missing more than 50% of my Affy matches. One is a predicted 4th-distant cousin, and the rest are all predicted 5th-distant cousins. I also had a Affy full 4th cousin who is now a 5th-distant cousin on Illumina.

I also manage another kit (182884) originally done under Affy, but I don't know the breakdown off the top of my head.

Judy

[gtc]

I.T. came back and says there are NO software bugs. There is a bug in yDNA reporting but that's different, not related to Family Finder, and its been announced to project admins.

Ah, seems my "little birdie" got their wires crossed in that regard.

Max is heading up everything on this issue now.

So management finally gets involved in the issue.

I believe I know what they will say and in my opinion it will be something about the tests being different and that Affy matches misisng do not mean they are invalid, just they have such small quantities of DNA they were not detected by Illumina, a test with more data points.

Small samples? I did fresh scrapes for FF when they could not use my existing samples. They got what they asked for in my case.

I reiterated we the customers need to see something "official" from FTDNA

Thank you for doing what FTDNA's management should have worked out by themselves.

Keep up the good work.