Fróðskaparrit - 01.01.1998, Side 7

ARVAEGINLEIKAKANNINGAR í FØROYUM
13
Table 2. The probability of segment sharing by
chance with two neighbouring markers not linked
to the disease in proportions out of 16
chromosomes. P(com) is the probability when the
most common allele at each locus makes the pair,
p(IBS) are the probabilities summedfor all possible
pairsfor the dijferent haplotypes.
Talva 2. Víst verður hvussu sannlíkt tað er, at eitt
felags petti við tveimum grannamørkum verður
arvað, mett sum tøl í mun til 16 kromosom. P(com)
er hvussu sannlíkt tað er, tá títtasti allelur gongur
sum par, p(IBS) er hvussu sannlíkt tað er, tá allir
møguleikar eru við fyri ymsar haplotypir.
p(com) p(IBS)
any 3 out of 16 0.23778 0.70104
any 4 out of 16 0.14720 0.24294
any 5 out of 16 0.06729 0.08386
any 6 out of 16 0.02350 0.02584
any 7 out of 16 0.00639 0.00666
any 8 out of 16 0.00137 0.00139
share a haplotype which is by chance iden-
tical by descent can be calculated from for-
mulas derived by Houwen et al. (1994)
(Table 2). A s shown even as little as 2 out
of 8 individuals sharing a haplotype identi-
cal by descent is a rare event which even a
genome wide scan only occurs a few times
for persons related six or more generations
ago.
When considering alleles at two neigh-
bouring markers and which alleles are in
linkeage equilibrium the probability of seg-
mentsharing just by chance is dependent on
the marker allele frequencies. For two
neighbouring markers with 6 alleles each
with frequencies of 0.4, 0.2, 0.15, 0.1 and
0.05 these probabilities are shown in Table
2.
It is crucial to choose which amount of
sharing that is needed before a chromoso-
mal area is considered iteresting. If the
treshhold is set too high the disease gene re-
gion may be missing and if the treshhold is
set too low to many false positive regions
will have be tested with additional markers.
Genetic analyses
DNA was prepared from whole blood, us-
ing a standard triton lysis, nuclear lysis pro-
tocol with sodium chloride/isopropanol
precipitation. DNA amplification was em-
ployed using a Perkin-Elmer thermocycler
with optimised reaction conditions for each
primer set and analysed on the ABI Prism
310 Genetic Analyzer.
As part of an ongoing genome wide scan
twelve markers were tested on chromo-
somes 16 and 18 (Table 3).
Results
Preliminary results are shown in Table 3.
On chromosome 16 at least 10 alleles were
shared for the most distal marker tested
D16S2622. Evidence of segment sharing
did not recieve support from the marker
proximal for D16S2622.
On chromosome 18 for marker D18S877
11 out of 16 alleles were shared while for
marker D18S541 13 out of 16 alleles were
shared. This did receive some support from
neighbouring markers.
Discussion
As the Faroese population descended from
Scandinavian and British ancestors, and
probably are founded from much fewer
people than Iceland, disease genes found in
this population are of great potential rele-
vance.