Raw data for both parental lines?

[Ann Turner]

You're correct that the maternal and paternal results aren't listed in separate columns.

However, the order isn't totally random -- the pairs of bases are actually listed alphabetically: AC, AG, AT, CG, CT, GT (and AA, CC, GG, TT). You won't see the pairs listed in reverse alpha order such as CA, TC, etc.



Elise

That is true for 23andMe, but FTDNA doesn't follow that convention exclusively. Here are the counts for different genotypes in one sample file:

AA 114896
AC 19570
AG 84550
AT 120
CA 0
CC 133889
CG 203
CT 0
GA 0
GC 194
GG 133243
GT 0
TA 109
TC 85435
TG 19526
TT 114124

[efgen]

Each position can be either, they are listed in random order.

You're correct that the maternal and paternal results aren't listed in separate columns.

However, the order isn't totally random -- the pairs of bases are actually listed alphabetically: AC, AG, AT, CG, CT, GT (and AA, CC, GG, TT). You won't see the pairs listed in reverse alpha order such as CA, TC, etc.



Elise

[prairielad]

Looking for insight into the raw data.

...................

Are long stretches of column 1 associated with one parent, and the same stretch of column 2 associated with the other parent?


Thanks,

Doug Price

In regards to Raw Data file, no column 1 is not associated with one parent and column 2 with other parent

meaning that all values on left belong to say father and right to mother. Each position can be either, they are listed in random order.

Would be nice if they would list them that way, but do not know the science on how they split and read each chromosome pair.

You would have to compare parent to child raw Data to seperate out the maternal/paternal chromosome values.(Phasing).
Or if not possible you can compare known relatives Raw Data to aid in determing which value goes to which.

edit (I asked simialer question in following thread)

http://forums.familytreedna.com/showthread.php?t=31245

[tomcat]

No, the results are un-phased, phasing being the term for the attribution of SNP's to one or another parental chromosome. So, you can't tell which match should be attributed to which parent unless the match also matches to a surname, or ethnicity or geography specific to one parent.

Looking for insight into the raw data.

The FamilyFinder test using the Illumina chip provides about 710,000 allele pairs (paternal and maternal alleles at each SNP locus).


Does the Illumina chip separately decode the maternal and the paternal chromsomes, i.e., in the raw data is one column of the alleles associated with one parent (even though it can't tell which parent it is) and the other column associated with the other parent?


Or, at the other extreme, is the allele pair at each locus in random order -- one comes from one parent and one form the other, but there's no associationm with the alleles at adjacent loci?

Are long stretches of column 1 associated with one parent, and the same stretch of column 2 associated with the other parent?


Thanks,

Doug Price