Error message when trying to upload 23me raw data

[dmc68]

Hey Everyone,
I am trying to upload my 23me raw data file. I uncompressed it as a text file and did the upload. It gave me an error message, so I called customer support. They told me to use a zip compression utility, so i downloaded the one they suggested. I knew it would not work, and it didnt.
I submitted a help desk report and they said 3-5 business days which is a bunch of crap.
There is one thing I noticed that was strange, they talk about submitting your 23me relative finder v3. This is not the name of the file so I have no clue what ftdna means by this.

If anyone has any suggestions on how I can resolve this issue I would greatly appreciate it.

Dave

[Ann Turner]

Hey Everyone,
I am trying to upload my 23me raw data file. I uncompressed it as a text file and did the upload. It gave me an error message, so I called customer support. They told me to use a zip compression utility, so i downloaded the one they suggested. I knew it would not work, and it didnt.
I submitted a help desk report and they said 3-5 business days which is a bunch of crap.
There is one thing I noticed that was strange, they talk about submitting your 23me relative finder v3. This is not the name of the file so I have no clue what ftdna means by this.

If anyone has any suggestions on how I can resolve this issue I would greatly appreciate it.

Dave

23andMe is now on v4 of the chip, and FTDNA doesn't accept that. There was a short period of time toward the end of 2013 when some people received v3 data and other people received v4. The file name or header doesn't tell you which version you have. One way to check is to see if you have results for this SNP. If you do (even if it is a no-call), then you have v4.

https://www.23andme.com/you/explorer/snp/?snp_name=i4001377

[dmc68]

So, if I get what you are saying, I cannot use my dna from 23me?

[Ann Turner]

So, if I get what you are saying, I cannot use my dna from 23me?

I can't tell without more information. Did you have a result for that SNP?

[dmc68]

I have no clue what an SNP is.

[MoberlyDrake]

Do you know what date you got your test results at 23AndMe? If it was this year, you have Version 4 and can't transfer.

[hansonrf]

So, if I get what you are saying, I cannot use my dna from 23me?

You can use it at GedMatch.com; just recognize some data will be 'imputed' by 23andMe and at least I can't tell you which. I think GedMatch just assumes data is data...

[Lincoln]

dmc68 - I think you need to go to the link Ann Turner provided if there is question about whether you are V3 or V4 at 23andMe. when I go to that link: https://www.23andme.com/you/explorer..._name=i4001377 , since I am V3, I get this answer from 23andMe:

no gene or snp found

[Ann Turner]

You can use it at GedMatch.com; just recognize some data will be 'imputed' by 23andMe and at least I can't tell you which. I think GedMatch just assumes data is data...

23andMe is not actually doing any imputation for the raw data download. GEDMatch makes the reasonable assumption that if you match a long run of consecutive SNPs from v4 data, you would also match the v3 SNPs in the same vicinity.

[hansonrf]

23andMe is not actually doing any imputation for the raw data download. GEDMatch makes the reasonable assumption that if you match a long run of consecutive SNPs from v4 data, you would also match the v3 SNPs in the same vicinity.

Good to know. Did they actually ever say that or is this a discovery you have made on your own? I'm a 23andMe customer and have found details like that hard to find. [I do believe you, though. I'm a fan.]

Sounds like GedMatch is doing 'imputation' of their own. They always have 'made positive assumptions' between kit types and to handle no-calls, etc., though not calling it imputation. [I never liked the use of that term anyway as it has a negative overtone...]

Do you, personally, think FTDNA's concern over v4 data is valid? I'm suspicious of both sides since both would peddle snake-oil if we'd buy it. GedMatch seems to handle it OK.

As a scientist [retired] I have great difficulty with calling much of this true science; a lot of smoke and mirrors; too much for real science. That said, it is intriguing and a marvelous hobby, and has led me to significant findings not otherwise possible.

[Ann Turner]

Good to know. Did they actually ever say that or is this a discovery you have made on your own? I'm a 23andMe customer and have found details like that hard to find. [I do believe you, though. I'm a fan.]

Sounds like GedMatch is doing 'imputation' of their own. They always have 'made positive assumptions' between kit types and to handle no-calls, etc., though not calling it imputation. [I never liked the use of that term anyway as it has a negative overtone...]

Do you, personally, think FTDNA's concern over v4 data is valid? I'm suspicious of both sides since both would peddle snake-oil if we'd buy it. GedMatch seems to handle it OK.

As a scientist [retired] I have great difficulty with calling much of this true science; a lot of smoke and mirrors; too much for real science. That said, it is intriguing and a marvelous hobby, and has led me to significant findings not otherwise possible.

Imputation means predicting the actual allele, based on how often it co-exists with alleles for neighboring SNPs ("linkage disequilibrium").

For some applications (e.g. medical), you do want to know the actual allele.

For other applications (e.g. segment matching), you only care if the allele is the same or different. If an unknown SNP (whether a no-call or not covered by the chip) falls in the midst of a long consecutive run of half-identical SNPs, it's a reasonably safe assumption that the unknown SNP would also match. I do think FTDNA could do some empirical testing within their own database by creating simulated records, replacing missing SNPs with no-calls.

I have a v4 test for comparison with my v2 results at 23andMe. I posted a very long message there, but we're not allowed to give URLs. Bottom line, the segment lengths averaged a tad longer with v4, about 0.1 cM. This is not unexpected, since a contradiction to the long consecutive run might not occur quite as quickly. But with very few exceptions, I neither gained nor lost matches within the 23andMe database.

[hosea49]

Comparisons at GedMatch

My older brother and I were both tested at 23andMe and uploaded to GedMatch. Our Maternal matches perfectly, but there seems to be completely different info for Paternal; we share 25.9%. The implication is pretty upsetting if we are indeed half brothers, but my question is do we assume this is correct or is there another way to verify? I am on FTDNA Family finder, Y chromosome tested to marker 111, but the program will not accept my brother's 23andMe raw Data, apparently they accept only the older V3. They did accept my Ancestry raw data.
Comments or suggestions from anyone would be appreciated.
Hosea

[Tenn4ever]

Hosea, is Gedmatch working for you this morning?

[benzden]

family finder v3 raw data file NOT!

So, finally re-accessed my nephew's 23andme account to obtain the genome-…zip file for his raw data. Can't upload it here because apparently it's a v4 versus v3 file - meaning our attempt to create a site here for our 16 generations of known genealogy for both paternal and maternal lines with at least two known males members of our family is going nowhere.

And, would I have to create another account for him here (instead of using mine for both of us) in order to not confuse our separately obtained DNA tests? (mine is only the Y-DNA12 - his, of course, must be Y-DNA23)

[MMaddi]

So, finally re-accessed my nephew's 23andme account to obtain the genome-…zip file for his raw data. Can't upload it here because apparently it's a v4 versus v3 file - meaning our attempt to create a site here for our 16 generations of known genealogy for both paternal and maternal lines with at least two known males members of our family is going nowhere.

Although you can't upload your nephew's 23andMe v4 raw data into FTDNA's Family Finder database, your nephew can order the Family Finder test directly at https://www.familytreedna.com/products.aspx. He'll have to send a DNA sample to FTDNA to get the test done.

And, would I have to create another account for him here (instead of using mine for both of us) in order to not confuse our separately obtained DNA tests? (mine is only the Y-DNA12 - his, of course, must be Y-DNA23)

That's correct. Each individual has a separate account for themselves at FTDNA. Your nephew would have to order a test kit from the page I linked to above, since he's a new customer for FTDNA and they don't have his DNA sample. Since you're an existing customer, you can order tests from within your existing account. But you and your nephew will need to have separate FTDNA accounts.