DYF399S1 is an interesting marker, but the alleles have a very complex structure that need some more clarification if they should be reported:

Gareth Henson has defined the structure of DYF399S1 as follows:
(GAAA)3AA(GAAA)A(GAAA)n

Gareth and I had a closer look at the structure after considering the newest ISFG guidelines:
(http://www.cstl.nist.gov/div831/strbase/pub_pres/ISFG_%20Y-STRupdate.pdf)

Considering the recommendations we looked up the published chimp sequences and found some parts of the flanking regions could also be variable and should contribute to the repeat numbers.

The new counting sheme is 6 repeat units longer and has the structure:

AAAAAAT-AAAAG-(AAAG)2-AAAAAG-AAAAG[G]-(AAAG)n-[AAAA]AAC

The bases in squared brackets can be deleted in some motifs.

We concluded that all AAAN blocks could have been normal AAAG repeats a long time ago. So the procedure would be counting all AAAN structures and summing up all additional bases behind the dot.

I'll try to explain this with all the motifs that are in the public human DNA sequence database of the Human Genome Project (HUGO). Every repeat motif is in a new lane and extra bases are labeled with a capital letter:


First structure


...gggttttcaccagtttgcataggtagagggaggccaaaagcccaacagg


AAAaaat
Aaaag
aaag
aaag
AAaaag
AaaagG
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
AAAaaac

26 full repeats (at least AAAN in each group)
11 extra bases (10 x A, 1 x G)


ttttacccttttgacagcatatgagacttctgggttcttttctcctgggtccaatcct
aagctgtccagtttaatgtttgggaaattaactcttccaaacttggaggatgcat
tgaagaggaatgtcccaaaacatgg...

302 bp total length of the PCR product


Second structure

...gggttttcaccagtttgcataggtagagggaggccaaaagcccaacagg

AAAaaat
Aaaag
aaag
aaag
AAaaag
Aaaag~
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaagAAC

23 full repeats (at least AAAN in each group)
10 extra bases (9 x A, 1 x C)


ttttacccttttgacagcatatgagacttctgggttcttttctcctgggtccaatcct
aagctgtccagtttaatgtttgggaaattaactcttccaaacttggaggatgcat
tgaagaggaatgtcccaaaacatgg...

289 bp total length of the PCR product


Third structure

...gggttttcaccagtttgcataggtagagggaggccaaaagcccaacagg

AAAaaat
Aaaag
aaag
aaag
AAaaag
Aaaag~
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
aaag
AAAaaac

24 full repeats (at least AAAN in each group)
10 extra bases (10 x A)

ttttacccttttgacagcatatgagacttctgggttcttttctcctgggtccaatcct
aagctgtccagtttaatgtttgggaaattaactcttccaaacttggaggatgcat
tgaagaggaatgtcccaaaacatgg...

293 bp total length of the PCR product


Remarks:
The motif of the repeat is 'AAAN' ('N' is the IUB code for 'any base').
If there would occur a motif AAAAAAAN (8 or more bases) then this would be broken up in two motifs (AAAA-AAAN).

The short form of allele reporting would only count the extra G in AAaaagG. HUGO would be 23-24-26.1
The full form of reporting would include all extra bases. HUGO would be 23.10-24.10-26.11

Sorry for the complicated nomenclature, but we have to be very careful if we want to report a complex marker like DYF399S1.