DYS464
extended test aka DYS464x plus some comments on DYF399
John McEwan
4th June 2006
Background
DYS464 is a multi-copy Y
chromosome STR, with typically 4 copies, but ranging from 0 to 7, and is
offered by several companies in particular FTDNA. The marker is quite difficult
to handle from an analysis perspective, because like other multi-copy repeats
the standard tests cannot distinguish the individual copies. Convention lists
them from lowest to highest. This makes comparison between diverged individuals
fraught because it is often not clear which copy has altered.
These tests are also subject to additional issues, because unlike normal
markers peak height has to be used to score whether they have one or more
copies of a particular allele. Certain combinations of alleles make this
difficult, especially if copy number also varies as well eg AABB and ABnullnull.
However, in practice the marker has been shown to be extremely useful
from a genealogical perspective, both in distinguishing families, and for
predicting haplogroup.
More recently, an extended test has been developed DNA-Fingerprint that can in
some circumstances distinguish between some of the alleles. This
is possible because there has been a G->C SNP mutation identified, which
“tags” the alleles where it is present. However, this test also
identifies a further process that affects variability in this marker called gene conversion or recombinatorial loss of heterozygosity or RecLOH for short. This happens in part because sections of the Y
chromosome are arranged and duplicated in palindromes aka inverted
repeats so they can repair or delete themselves even though only one copy of
the DNA is available. The genomic organization of DYS464 and DYF399 in the
AZFc region is shown in Figure 1
below taken from Henson (2005).
The mutation rate for SNPs is about once every 50,000,000 generations
and for STRs about once every 500 generations: RecLOH events have an intermediate frequency between these events.
The paper by Henson also summarizes that these RecLOH events and deletions often affect multiple markers and in
this case both DYF399 and DYS464.
Variability of DYS464x in the
R haplogroup
This section expands and updates a Genealogy DNA listserver message
by Gareth Henson in November 2005. Most non R1b1c individuals have 4 DYS464 alleles and the SNP is G i.e. GGGG. At some point a G->C mutation occurred and through RecLOH it was converted to CCGG and then a further duplication to CCCG. The CCCG form is the one most commonly observed in R1b1c individuals. Thus an initial mutation and then two further
steps were required. This is almost certainly a unique event. However, the RecLOH process means that additional
variability is observed. This makes the test a “quazi-SNP” i.e. less variable
than an STR, but more variable than a SNP and it has value both for exploring
deep ancestry and as a genealogical marker.
Exactly when the DYS464 GGGG
form changed to CCCG form is not
known, but public results from Ymatch show the GGGG form for haplogroup A
(n=1), E (2), G (3), J (1 GGGG, 1 GGnullnull), I (4), R1a (1). The majority of the R1b1c records are of the CCCG form. There is also one R1b1* (P25+, M269-) that has the GGGG form (grey see table 1 below). A M269+, S21-, S28- individual with the GGGG form has also been identified (see R1b SNP table). This suggests that perhaps this
change most likely happened soon after the M269+
mutation, and before the S21 and S28 mutations as it is unlikely to have
had 2 RecLOH events to revert to the
GGGG form. However, this hypothesis
requires further results to be confirmed.
Deletion events in DYS464 may
also affect adjacent markers as well especially DYF399S1, DYF371, DYF385, DYF399 and DYF408. Table 1 is an edited extract of R1b from Ymatch with
only the multiple copy markers retained that have been scored for DYS464x. Although care is needed as some entries may
be mis-scored (highlighted pink, all with double peaks which are hard to
reconcile) it is generally observed that when only two alleles are DYS464 are present: only one allele of DYS724, two alleles of DYS725, three alleles of DYF371, one allele of DYF401 and 3 alleles of DYF408 are also present. Note that CCnullnull (n=3) and CGnullnull (n=2) seem equally prevalent
out of 80 R1b DYS464x records
available. They also identify that these deletions have happened multiple
times. Insufficient records have been collected as yet to identify how
diagnostic these markers are, but any R1b DYS464
results that are of the form XXXX are prime candidates for deletion testing and
at least DYS464x, DYF399 and DYS371 should be tested together to
validate the results. There is equivocal evidence that some of these myriad of
partial deletions may reduce male fertility in a minority of cases, but it is
clear that deletion of the whole region results in infertility (see Vogt, 2005
for a review and Carvalho et al 2006 for a more recent reference).
The other common event is a CCCG
-> CCGG or CCCG -> CCCC RecLOH with 7 out of 80 and 1 out of 80
respectively. These events have been observed separately in S21+ and S21- individuals (see R1b SNP table) again
suggesting multiple events.
In summary, the DYS464x test
and the associated DYF399 test
provide valuable information especially within R1b and are identifying useful variants for reconciling close
matches and also potentially for deeper ancestry.
Table 1. Ymatch R1b
records for multiple copy markers highlighted examples of probable RecLOH
events and deletions.
References
Carvalho
CM, Zuccherato LW, Bastos-Rodrigues L, Santos FR, Pena SD. 2006. No association
found between gr/gr deletions and infertility in Brazilian males.
Mol Hum Reprod. 12:269-73
DNA-Fingerprint
explanation of RecLOH
DNA-Fingerprint
sample results from DYS464x
Henson, G.L. 2005 DYF399S1: A
Unique Three-Copy Short Tandem Repeat on the Human Y Chromosome. Journal of
Genetic Genealogy 1:8-11
Vogt
PH. 2005. AZF deletions and Y chromosomal haplogroups: history and update
based on sequence. Hum Reprod Update. 11(4):319-36
Wikipedia
repair of the Y chromosome